observations
190 rows · 20 columnsEvery row exactly as stored. No scoring, no gate composition, no coverage, no derived fields; NULL is shown as NULL. If a dossier disagrees with this page, one of them is lying and this one is not computing anything.
| # | id uuid | entity_id text | subject text | value jsonb | unit text | source_url text | publisher text | effective_at date | retrieved_at timestamp with time zone | source_tier source_tier | verification_status verification_status | collection_method collection_method | statute_source_url text | note text | created_by text | last_edited_by text | revision_id uuid | prior_revision_id uuid | changed_at timestamp with time zone | instrument_grade instrument_grade |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | c44aef6c-200f-485f-acc5-d5799a7d178b | ae-dubai-rove-hotel | capacity | {"rooms":420} | rooms | https://www.myboutiquehotel.com/en/boutique-hotels-dubai/rove-downtown-dubai.html | MyBoutiqueHotel (hotel directory listing), corroborated by ZenHotels | 2026-07-24 | 2026-07-24T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | DOCUMENT_EXTRACTION | NULL | 420 ROOMS IS MEASURED AND CORROBORATED ACROSS INDEPENDENT DIRECTORIES. THE BED COUNT IS A POLICY, NOT A FACT (D-040). SOURCE QUOTES, verified verbatim by an adversarial verifier that re-fetched every page (D-032): MyBoutiqueHotel reads 'Rove Downtown … Dubai - Downtown 420 rooms' plus 'The hotel provides spacious 26sqm rooms of which 30 percent are interconnecting'. The verifier independently re-fetched a SECOND, distinct-family directory — ZenHotels (HTTP 200) — which states '420 rooms, 14 floors' (and 'Year of construction 2016'), agreeing exactly. The OPERATOR's own page (rovehotels.com, HTTP 200) describes the hotel but does NOT render a room count, and Emaar's domain Cloudflare-403'd the fetcher, so the count rests on the two independent directories rather than the operator — enough for CORROBORATED, and nothing that rendered gave a conflicting number. WHY THIS IS A {rooms} CLAIM AND BEDS ARE NON-DERIVABLE (but for a different reason than the King/Twin hotels). Rove's stock is UNIFORM: every guest room is the same 26 m² 'Rover Room' (variants Rover Room, Burj View, Gamer Cave), configured 1 King bed + 1 sofa bed, and the operator's fine print states 'the sofa bed is suitable for a child and not for an adult.' So there is no unpublished King/Twin split to bound — but adult berths per room are ~1 King (a couple sharing) and the sofa is not an adult berth, so a single 'total beds' number is still config-dependent, not a building fact. Beds per room stays a scenario policy (bedsPerRoom): at 2.0 (D-059) this yields 840, at the old 1.0 floor 420 — either clears the 150 requirement without over-claiming. EFFECTIVE DATE: the directory listings assert present inventory and the operator page is undated, so effectiveAt = the retrieval date. Room stock at an operating Emaar-brand hotel (opened 2016) is not volatile; a dated operator confirmation would still settle it. TIER: COMMERCIAL_LISTING because the count comes from booking directories, not the operator's own statement or an official register. | seed | seed | 673724f8-d18d-4264-a58a-7761fb84738b | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 2 | 789da538-586f-420d-9418-9bccd1380d91 | ae-dubai-rove-hotel | property-monthly-cost | UNRESOLVED | NULL | https://www.rovehotels.com/en/hotels/rove-downtown/ | Rove Hotels / Emaar Hospitality (operator) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | A REAL MONTHLY RATE EXISTS AND IS RECORDED, BUT IT IS NOT A MEMBER-MONTH CONTRACT COST. The operator page advertises, verbatim (verifier-confirmed, HTTP 200): 'Monthly Hotel Rates — Monthly rates from just AED 3,499 per month, all included.' This is the FIRST published monthly figure in the property field — Rixos Almaty and Forest City Golf Hotel published no rate at all — and it is recorded here for that reason. It is HELD at NEEDS_EVIDENCE rather than scored because it does not answer the criterion, which is US$/member-month for a ~300-member multi-year node: (a) it is a 'from' (starting) promotional rate, not a firm price; (b) it is for a SINGLE Rover Room on a transient monthly stay, not a negotiated bulk multi-year contract; (c) it is per room, not per member, so converting it (÷2 adults ≈ AED 1,750 ≈ US$476/member-month at ~0.272 AED→USD) would stack a from-price, an occupancy assumption and an FX assumption into a number the product would then rank — the exact over-claim D-026 warns against. A firm bulk quote is a sales-desk question — intake, not research (DATA-SOURCES.md §116). | seed | seed | b1cc74d0-8655-4837-a9a1-f5f056acefe8 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 3 | 58db2377-f55c-4c3f-910a-c97342f24505 | ae-dubai-rove-hotel | workstations | UNRESOLVED | NULL | https://www.rovehotels.com/en/hotels/rove-downtown/ | Rove Hotels / Emaar Hospitality (operator) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | CO-WORKING IS ADVERTISED BUT UNCOUNTED. The operator page states, verbatim (verifier-confirmed, HTTP 200): 'Co-working spaces that keep you inspired, focused and moving' and 'Your Space to Work, Meet & Hang Out', plus a business centre — materially more than Rixos Almaty, which published only meeting/ballroom event space. But the `workstations` criterion needs a COUNT of usable workstations, and no desk/seat count is published anywhere. Recorded so the finding is visible, held NEEDS_EVIDENCE for want of a number; a count is a site-visit / operator question (intake, not research). | seed | seed | 4518b8bd-00fc-419b-b09a-ad727bcb0579 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 4 | 886a5358-a182-4641-9488-e958fe58a7b1 | ae-dubai-rove-hotel | availability | UNRESOLVED | NULL | https://www.rovehotels.com/en/hotels/rove-downtown/ | Rove Hotels / Emaar Hospitality (operator) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. The operator sells nightly and monthly-stay rooms but states nothing about a whole-property buyout or block booking for a multi-year group. Silence is not permission. D-026 SPLITS THIS QUESTION AND THE SCHEMA DOES NOT: whole-property buyout and block booking are different instruments with different answers, and this gate has one field for both. That a monthly-stay product exists (see property-monthly-cost) is NOT a multi-year commitment for 300 rooms. An availability date is obtainable only from the sales desk — intake, not research (DATA-SOURCES.md §116). | seed | seed | 63b64b98-632d-4870-b967-2d34a897a1a8 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 5 | 8ddd8b4b-1c0e-4203-8b23-e45d31600de5 | ae-dubai-rove-hotel | title-instance | UNRESOLVED | NULL | https://www.rovehotels.com/en/hotels/rove-downtown/ | Rove Hotels / Emaar Hospitality (operator) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. Dubai Land Department title records are not published free-to-fetch, so there is no automatable source (DATA-SOURCES.md §194) and none was invented. THE INFERENCE DELIBERATELY REFUSED: 'a hotel operates lawfully as a hotel, therefore its title supports our use' does not follow. 300 people in 180+ day residential occupancy, with teaching on site, is a different use class from transient accommodation, and nothing establishes it is permitted. The COUNTRY-level foreign-title-statute gate resolves at UAE and is a separate question from this instance-level gate (D-011). | seed | seed | f567406c-bd55-437e-98f7-f59a87c88fd7 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 6 | 87fae1b8-96cd-4d00-9e00-0056ebccdf2e | id-bali-denpasar-aston-hotel | capacity | {"rooms":211} | rooms | https://www.cvent.com/venues/bali/hotel/aston-denpasar-hotel-convention-center/venue-46292954-86ce-4906-903c-1ad7918a37c2 | Cvent (MICE venue directory), corroborated by Trip.com | 2026-07-24 | 2026-07-24T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | DOCUMENT_EXTRACTION | NULL | 211 GUEST ROOMS IS THE SOURCED, CORROBORATED FIGURE; 258 IS THE 'ROOMS AND SUITES' TOTAL AND IS DOCUMENTED, NOT RECORDED AS THE VALUE. THE BED COUNT IS A POLICY, NOT A FACT (D-040). THE CONFLICT AND ITS RESOLUTION (D-032). The Cvent venue page (HTTP 200) states BOTH 'The Hotel & Convention Center offers over 258 rooms and suites' (prose) AND a structured field 'Guest Rooms 211' / 'Total guest rooms 211'. An adversarial verifier that re-fetched independently reproduced the IDENTICAL split on a second, distinct-family source — Trip.com (HTTP 200): structured 'Number of Rooms: 211' (with 'Opened: 2008') alongside prose 'one of the 258 air-conditioned rooms'. So 211 is the standard-guest-ROOMS count (structured, on two independent renderers) and 258 is the total-keys figure INCLUDING SUITES (marketing prose, on both). The 211 + ~47-suite decomposition is arithmetic inference (258−211) and is NOT quoted by any source, so it is not asserted. Per the reference-integrity principle (prefer weaker-but-right, never drop a source to simplify), the VALUE recorded is the explicitly-sourced 211; 258 is documented here as the total-accommodation prose figure. The property demonstrably has multiple suite types (Aston/Junior suites per the listings), consistent with 258 total keys. THE GATE IS ROBUST TO THE CONFLICT. The capacity gate needs >=150 beds; at bedsPerRoom 2.0 (D-059) both 211 (->422) and 258 (->516) clear it, and even at the old 1.0 floor both clear 150. So the capacity PASS never depended on which figure is used — only the stored record does, and it is the conservative sourced one. BEDS ARE POLICY-BOUND (D-040), as at every hotel: the room/suite mix is not a bed count; beds per room lives on the scenario (bedsPerRoom), and the gate names the conversion in its reason. TIER: COMMERCIAL_LISTING because the count comes from booking/venue directories (Cvent, Trip.com), not the operator's own statement (astonhotelsinternational.com CloudFront-403'd the fetcher) or an official register. effectiveAt = retrieval date; the directories assert present inventory and the operator page was unreachable to date it. Corroborating context (verifier, not the value): the hotel opened 2008 and carries one of Bali's largest ballrooms (Grand Ballroom ~13,670 sq ft, theater 2,000, 12 meeting rooms) — event space, not member workstations. | seed | seed | ee345629-9c82-4a51-94d6-607cd100deb6 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 7 | d592a257-a7e5-4f05-b75c-3495ba4f7c8c | id-bali-denpasar-aston-hotel | availability | UNRESOLVED | NULL | https://www.cvent.com/venues/bali/hotel/aston-denpasar-hotel-convention-center/venue-46292954-86ce-4906-903c-1ad7918a37c2 | Cvent (MICE venue directory) | 2026-07-24 | 2026-07-24T00:00:00.000Z | COMMERCIAL_LISTING | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. No source addresses a whole-property buyout or a multi-year block for a ~300-member group; the venue directory lists event bookings, not residential accommodation contracts. Silence is not permission. D-026 SPLITS THIS QUESTION AND THE SCHEMA DOES NOT: whole-property buyout and block booking are different instruments with different answers, and this gate has one field for both. An availability date is obtainable only from the sales desk — intake, not research (DATA-SOURCES.md §116). | seed | seed | 5098d831-3408-458e-b61a-05722a924f0f | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 8 | 7bd84a3b-5d36-4512-a90b-89454b3edf36 | id-bali-denpasar-aston-hotel | title-instance | UNRESOLVED | NULL | https://www.cvent.com/venues/bali/hotel/aston-denpasar-hotel-convention-center/venue-46292954-86ce-4906-903c-1ad7918a37c2 | Cvent (MICE venue directory) | 2026-07-24 | 2026-07-24T00:00:00.000Z | COMMERCIAL_LISTING | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. Indonesian land records (BPN) are not published free-to-fetch, so there is no automatable source (DATA-SOURCES.md §194) and none was invented. THE INFERENCE DELIBERATELY REFUSED: 'a hotel operates lawfully as a hotel, therefore its title supports our use' does not follow. 300 people in 180+ day residential occupancy, with teaching on site, is a different use class from transient accommodation, and nothing establishes it is permitted. The COUNTRY-level foreign-title-statute gate resolves at Indonesia and is a separate question from this instance-level gate (D-011). | seed | seed | 2c8df050-ff7c-4666-9fad-35892f493b84 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 9 | c60ed75f-527c-4a6e-9f8c-0039a9686f5e | kz-almaty-rixos-hotel | capacity | {"rooms":237} | rooms | https://all.accor.com/hotel/B1N9/index.en.shtml | Accor (operator, ALL property page for Rixos Almaty) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | CORROBORATED | DOCUMENT_EXTRACTION | NULL | 237 ROOMS IS MEASURED AND CORROBORATED ACROSS INDEPENDENT SOURCE FAMILIES. THE BED COUNT IS NOT, AND IS NOT DERIVABLE (D-040). SOURCE QUOTE, verified verbatim by an adversarial verifier that re-fetched every page (D-032): the Accor ALL property page reads 'Its 237 rooms and suites, many with views of the Pine Park, are among the largest in the city, with exquisite design.' The verifier independently re-fetched a SECOND Accor surface (allinclusive-collection.com, HTTP 200, identical string) AND a THIRD, NON-ACCOR source — Cvent's venue record (cvent.com/venues/almaty/hotel/rixos-almaty-hotel, HTTP 200) — which states 'Its 237 rooms and 7 meeting rooms…' and a venue block 'Guest Rooms 237' / 'Total guest rooms 237'. A distinct third-party MICE database agreeing at 237 is what lifts this to CORROBORATED, where Forest City Golf Hotel stood UNVERIFIED on its operator page alone (its second source, trip.com, bot-blocked the verifier). trip.com/TripAdvisor (HTTP 403) and Travel Weekly (timeout) refused this fetcher and are counted neither way; nothing that rendered gave a conflicting number — the 151–295 spread seen on Rahat Palace did not appear here. WHY THIS IS A {rooms} CLAIM AND NOT A BED COUNT. The operator sells 'Deluxe/Superior King Room' (1 King bed) and 'Deluxe/Superior Twin Room' (2 Twin beds) as separate 32 m² products, but publishes NO count of how many of the 237 are King versus Twin — Cvent gives only the aggregate 237. So the King/Twin mix is unknown and beds are bounded only, not derivable, from any source. Beds per room is a policy about how we intend to house members, not a fact about this building; it lives on the scenario (bedsPerRoom) and the capacity gate names the conversion in its reason. At the shipped policy of 2.0 (D-059) this yields 474 beds; even at the old 1.0 floor it is 237 — either way it clears the 150 requirement without over-claiming. EFFECTIVE DATE IS THE WEAKEST PART OF THIS ROW, exactly as at Forest City. The Accor page is UNDATED; its only temporal markers are '© 2024'/'© 2019' footers. effectiveAt is set to the retrieval date because a live sales page asserts present inventory and two further live pages state the same 237 the same day — but if the operator copy is a 2024 artefact this figure could age past the 365-day window. A dated source or an operator confirmation would settle it. (Room stock at a branded, operating 5-star is less volatile than at Forest City, which carries publicised occupancy pressure; the caution is retained but is weaker here.) | seed | seed | 79bf7ec5-1b0c-4b8b-abf7-7b19991d846a | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 10 | 4497f82d-ab16-47e9-a19f-36da8352c7ea | kz-almaty-rixos-hotel | availability | UNRESOLVED | NULL | https://all.accor.com/hotel/B1N9/index.en.shtml | Accor (operator, ALL property page for Rixos Almaty) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. The operator page sells nightly rooms and lists meeting/event facilities, but states nothing about a whole-property buyout, block booking, or a multi-year group contract. Silence is not permission: the absence of a stated restriction establishes nothing. D-026 SPLITS THIS QUESTION AND THE SCHEMA DOES NOT: whole-property buyout and block booking are different instruments with different answers, and this gate has one field for both. No source addresses either. A booking engine showing free rooms would not have established availability and was not treated as if it could — free inventory on a date is not a multi-year commitment. An availability date is obtainable only from the operator's sales desk (+7 727 300 3300, HB1N9-RE@accor.com), which is intake, not research (DATA-SOURCES.md §116). | seed | seed | c531f94e-7865-4907-9fa3-56421087b39e | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 11 | d3411941-0277-4801-b4b4-d6bc66a04826 | kz-almaty-rixos-hotel | title-instance | UNRESOLVED | NULL | https://all.accor.com/hotel/B1N9/index.en.shtml | Accor (operator, ALL property page for Rixos Almaty) | 2026-07-24 | 2026-07-24T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. Kazakhstan land-registry records are not published free-to-fetch, so there is no automatable source (DATA-SOURCES.md §194) and none was invented. THE INFERENCE DELIBERATELY REFUSED: 'a hotel operates lawfully as a hotel, therefore its title supports our use' does not follow. 300 people in 180+ day residential occupancy, with teaching on site, is a different use class from transient accommodation, and nothing establishes it is permitted. The COUNTRY-level foreign-title-statute gate resolves at Kazakhstan and is a separate question from this instance-level gate (D-011). | seed | seed | 51792dc7-37f4-4d9e-9231-8d7dae242137 | NULL | 2026-07-25T03:40:11.592Z | NULL |
| 12 | 6f1681b9-e291-4b25-8571-b71d13f9739c | my-jhr-forest-city-golf-hotel | title-instance | UNRESOLVED | NULL | https://fcgolfhotel.com/about-forest-city-phoenix-golf-hotel-johor-bahru | Country Garden Pacificview Sdn. Bhd. (operator) | 2026-07-17 | 2026-07-17T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. Malaysian land registry (PTG Johor) records are not published free-to-fetch, so there is no automatable source (DATA-SOURCES.md §194) and none was invented. THE INFERENCE DELIBERATELY REFUSED: 'a hotel operates lawfully as a hotel, therefore its title supports our use' does not follow. 300 people in 180+ day residential occupancy, with teaching on site, is a different use class from transient accommodation, and nothing establishes it is permitted. Forest City's reclaimed-land title history and its Special Financial Zone designation make this specifically unsafe to guess at. Note the country gate above it: foreign-title-statute resolves at COUNTRY and this instance-level gate is a separate question (D-011). | seed | seed | 8318eec7-40d5-4095-a90d-5b734effb937 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 13 | 25ff6f70-5b2e-4105-9d91-1d5ad3abbfd7 | uz-tashkent-city | air-quality | 34.8 | PM2.5 ug/m3 | https://api.openaq.org/v3/sensors/25916/years?limit=100 | StateAir Tashkent (US Diplomatic Post: Tashkent), via OpenAQ v3 API | 2024-12-31 | 2026-07-22T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | OFFICIAL_API | NULL | THE KEYED OPENAQ PASS, 2026-07-22 — the station-measured, differently-licensed comparator the modelled 38.8 row's own note asked for; BOTH ROWS STAND (this one, sorting first, is the scored one; the CAMx row remains the population-weighted context). LICENCE — STATED FIRST BECAUSE IT IS THE ONE OPEN QUESTION, OWNER RULING OWED: OpenAQ states NO licence for this location — location 8881 reads verbatim '"licenses":null', re-found three times by the adversarial verifier, and the provider/instrument records carry no licence field either. That is 'the API states no licence', NOT 'the data is unlicensed': the Almaty sibling station (location 8876, the same US-government data through the newer AirNow pipeline) carries 'US Public Domain' (licence id 33, commercialUseAllowed:true), so the null here is a legacy-StateAir-pipeline metadata gap — recorded as owner-ruling CONTEXT only, never as a licence claim for 8881. MONITOR GRADE: reference/government monitor — 'isMonitor":true', instrument 'Government Monitor', the US Embassy StateAir monitor, 7,993 m from the uz-tashkent-city centroid; the only other PM2.5 source within 25 km is a low-cost AirGradient with no complete year. WHAT THIS VALUE IS: the OpenAQ v3 yearly rollup of hourly PM2.5 at that single monitor for calendar 2024 — API-stated value 34.8, underlying avg 34.76265182186235, coverage 8761 of 8784 expected hourly observations = 99.7% (API-rounded 100.0); the rollup carries 'hasFlags":true'. WHAT IT IS NOT: not population-weighted and not city-wide (the modelled 38.8 is that construct, different method and vintage — the two are NOT comparable beyond 'same city'); not a current figure — the feed is DEAD since 2025-03-04 (2025 sits at 17% coverage and its partial rollup must never be quoted; the shutdown is press-corroborated as the US Department of State ending its overseas air-monitoring programme that same day, and the sensor's post-mortem 'latest' datum reads -999.0). TRAP, RECORDED SO NOBODY CONFLATES IT: this station's 2023 annual mean is coincidentally ALSO 38.8 — the same digits as the modelled CAMx figure, by accident. 2024 is the last and best complete year this monitor will ever produce. VERIFIER: CONFIRMED — 8 CONFIRMED / 0 CORRECTED / 0 REFUTED / 1 UNVERIFIED (a stateair.net wall, not load-bearing), every figure re-fetched verbatim with AUTH (evidence/openaq-air/{tashkent-claims,tashkent-verdicts}.md). | seed | seed | 21f3f0cd-b583-4cd7-b6c1-6a072d4d9a11 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 14 | 5b159e08-564c-4ecd-aa2d-c0819ab14911 | uz-tashkent-city | air-quality | 38.8 | PM2.5 ug/m3 | https://documents1.worldbank.org/curated/en/099061424075531584/txt/P1771081ec9e93071181d3140144918d6cd.txt | World Bank, in collaboration with the Ministry of Ecology, Environmental Protection and Climate Change of Uzbekistan | 2024-07-29 | 2026-07-20T00:00:00.000Z | MULTILATERAL | CORROBORATED | DOCUMENT_EXTRACTION | NULL | LICENCE — OWNER RULING 2026-07-22 (D-048), STATED BEFORE THE VALUE BECAUSE IT TRAVELS WITH IT: this source is 'CC BY-NC 3.0 IGO'. NonCommercial is asserted in two independent places, both retrieved: the Open Knowledge Repository citation block (https://openknowledge.worldbank.org/entities/publication/3fd0bf55-595c-481f-bd9b-23b40b9c55cc, HTTP 200) reads VERBATIM 'World Bank. 2024. Air Quality Assessment for Tashkent and the Roadmap for Air Quality Management Improvement in Uzbekistan. © World Bank. http://hdl.handle.net/10986/41969 License: CC BY-NC 3.0 IGO.', and the report's own front matter reads VERBATIM 'this work may be reproduced, in whole or in part, for non-commercial purposes as long as full attribution to this work is given.' The owner ruled to LAND the value WITH the restriction recorded here, on the reading that Atlas is a free, open-source, public product today. THE RESTRICTION DOES NOT EXPIRE: if this product ever takes money, this row must be re-ruled, and a keyed OpenAQ pull is the obvious differently-licensed comparator. VERIFIER: CONFIRMED (evidence/tashkent-city/verifier-verdicts.md §1) — the verifier fetched the 440,535-character full text itself and found the literal string '38.8' THREE times, in the three places named below. SOURCED FROM THE REPORT'S OWN FULL TEXT, not from a search summary and not from a commercial AQI mirror — and that distinction is the whole story here: a WebSearch summary asserted 38.8 to the World Bank, but every reachable secondary rendering (the worldbank.org publication page, the CCAC resource page, the gov.uz press release) and Wikipedia's Tashkent 'Ecology' section carry only a RATIO ('more than six times the WHO recommended level (5 μg/m3)') and NO µg/m3 figure at all. Anyone recording 38.8 without opening the primary would have been recording a search summary. VERBATIM occurrences, all re-found by the verifier: 'The WHO's annual average PM 2.5 guideline is 5 µg/m 3 , whereas the annual population-weighted average PM 2.5 concentration estimated for Tashkent is 38.8 µg/m 3 — over seven times above the WHO guideline.'; 'the population-weighted PM 2.5 annual average concentration for Tashkent that equaled 38.8 µg/m 3'; 'The attributable mortality for Tashkent caused by an annual average population-weighted PM 2.5 concentration of 38.8 µg/m 3 is 3,042 premature deaths per year.' WHAT THIS VALUE IS AND IS NOT — the bounding caveat: it is a MODELLED, POPULATION-WEIGHTED annual mean, not a single station's measured annual mean. VERBATIM on method: 'The Comprehensive Air Quality Model with extensions (CAMx) modeling system was used to simulate the dispersion of PM 2.5 pollution over the airshed', at 'a spatial resolution of approximately 1 by 1 km and a temporal resolution of 1 hour', built on 'an emission inventory for the key PM 2.5 emission sources'. Monitoring inputs are US Embassy and Uzhydromet station data 2018-2022. INTERNAL DISCREPANCY, reproduced exactly by the verifier: the report's Executive Summary says 'over six times' while the body says 'over seven times'; 38.8/5 = 7.76, so the body is self-consistent and the summaries understate. PUBLISHER WORDING CORRECTED per the verifier: the OKR Author field is 'World Bank' alone and the report's acknowledgements say the work was produced 'in collaboration with' the Ministry — the researcher's 'jointly with' over-attributed co-authorship and has been changed. effectiveAt = the PUBLICATION date (2024-07-29 per the OKR record), NOT a measurement year: the text states no single modelled base year. OPENAQ, the route this project normally uses, was walled and that is recorded as a fetch fact: GET api.openaq.org/v3/locations returned HTTP 401 '{"message": "Unauthorized. A valid API key must be provided in the X-API-Key header."}' for both agents. This value does NOT close the OpenAQ gap — a keyed checker should still pull the station-measured annual mean for comparison against this modelled figure. | seed | seed | 79c0d4bf-c3a2-4b4f-8a88-33925edaaa95 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 15 | 62b5f755-a004-400f-9868-63f7646e07d3 | uz-tashkent-city | sunshine-hours | 2755.6 | hours/yr | https://en.wikipedia.org/wiki/Tashkent | Wikimedia Foundation, relaying Pogoda.ru.net and NOAA (National Oceanic and Atmospheric Administration) 'Tashkent Climate Normals 1991-2020' | 2020-12-31 | 2026-07-20T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | TERTIARY RELAY, FLAGGED — and the verifier named this 'the value I would overturn first' (evidence/tashkent-city/verifier-verdicts.md, spot-check). CONFIRMED as what it says it is: the verifier re-fetched the page (HTTP 200), found the row verbatim, and recomputed the sum independently. VERBATIM from the table headed 'Climate data for Tashkent (1991-2020, extremes 1867-present)': 'Mean monthly sunshine hours 104.7 119.4 169.2 222.7 303.0 352.8 386.5 353.4 283.8 220.4 135.0 104.7 2,755.6'. The 12 monthly values sum to exactly 2,755.6. TWO CAVEATS THAT MUST NOT BE SKIPPED, both re-confirmed by the verifier: (1) PROVENANCE IS AMBIGUOUS — the table cites 'Source 1: Pogoda.ru.net' and 'Source 2: NOAA' for one table and does NOT attribute the sunshine row to either, so it cannot be asserted that 2,755.6 is NOAA's figure. It must NOT be promoted to a NOAA-sourced value by any later pass. (2) POSSIBLE TRANSCRIPTION ARTEFACT — January and December are listed as the IDENTICAL value 104.7, a recognised signature of a copy error in relayed climate tables; if it is an error the annual total is wrong. The verifier found a THIRD wobble in the same article: the body says an 'altitude of 500 m (1,600 ft)' while its own infobox says '455 m'. BOTH PRIMARIES ARE WALLED AND BOTH WALLS WERE REPRODUCED: NOAA 'Tashkent_38457.csv' (visible in the HTTP 200 directory listing at 4.7K) fails with 'page.goto: Download is starting' — the verifier also tested a sibling file and got the identical failure, proving a general file-download limitation of the headless fetcher, not anything about Tashkent; and pogoda.ru.net returns DNS_FAILED / ERR_NAME_NOT_RESOLVED, i.e. the host is gone. NO competing number is recorded: WebSearch surfaced wildly divergent annual totals from commercial climate sites, and a search summary is not a retrieval. RESOLVING CHECK, cheap: retrieve Tashkent_38457.csv with a client that accepts a file download and confirm both the annual total and the Jan/Dec values. D-049 DATING CONVENTION: effectiveAt = 2020-12-31, the END of the 1991-2020 normal period. The researcher recorded 1991-01-01 (the period START); it is moved here so that this row and Almaty's — the same WMO normal period, the same criterion — decay identically instead of differing by 30 years of freshness purely from a convention choice. Do not read effectiveAt as a measurement date. | seed | seed | b1a8d49b-aac5-4820-9894-38007e7135b2 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 16 | 9f7bee31-0baa-48f0-b35a-803ba9281dc2 | ae | foreign-title-statute | true | NULL | https://dlp.dubai.gov.ae/Legislation%20Reference/2006/Law%20No.%20(7)%20of%202006.html | Dubai Land Department / Dubai Legislation (dlp.dubai.gov.ae) | 2006-03-14 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | ANALYST_VERIFIED | DOCUMENT_EXTRACTION | NULL | TRUE, with an important structural caveat the engine should carry. The gate asks about NATIONAL statute; the UAE has no single federal statute governing foreign real-property ownership — property is devolved to the individual emirates. The sourced instrument is Dubai Law No. (7) of 2006, Article 4, which is EMIRATE-level. It establishes that a foreign entity may hold (a) freehold without time restriction, or (b) usufruct/leasehold up to 99 years, but ONLY in areas determined by the Ruler and subject to the Ruler's approval — i.e. the right exists as a matter of statute but is territorially bounded (designated/freehold zones) and discretionary as to approval. For a leased Network School campus the 99-year leasehold limb is the operative one and is comfortably sufficient. Scored true because the statute affirmatively grants the right to foreign holders; a reviewer wanting a stricter reading could argue the Ruler's-approval condition makes it conditional rather than statutory. Other emirates (Abu Dhabi, RAK, Sharjah) have their own and DIFFERENT regimes which I did not research — if the site is not in Dubai this observation does not transfer. SOURCE QUOTE, verbatim from the page on the retrieval date: “Subject to the approval of the Ruler, non-UAE nationals may, in certain areas determined by the Ruler, be granted the following rights: a. Freehold ownership of Real Property without time restrictions; and b. Usufruct or leasehold over Real Property for a period not exceeding ninety-nine (99) years.” | seed | seed | 04152b27-8892-48eb-ab07-72a4c80a95c6 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 17 | c9df3b43-f15c-4a74-af89-a0c9b7f1a87d | ae | operating-licence | UNRESOLVED | NULL | https://u.ae/en/information-and-services/education/higher-education/Higher-education-regulatory-bodies | The Official Platform of the UAE Government (u.ae) | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | I found the licensing REGIME but not the foreign-eligibility RULE, which is what the gate actually asks. Established: a licensing regime exists and is tiered — the federal Commission for Academic Accreditation (CAA, under MoHESR) licenses institutions and accredits programmes; KHDA licenses/oversees private institutions in Dubai; ADEK does so in Abu Dhabi; RAK DOK took over private-education regulation in Ras Al Khaimah. NOT established, and the reason this is UNRESOLVED: no page I could fetch states whether a FOREIGN entity may itself hold such a licence, on what conditions, or whether a non-degree-granting 'training institute' (the closer analogue to Network School than a university) falls inside or outside the regime. The KHDA permit guide at https://web.khda.gov.ae/en/Guides/Education-Providers/Permit-Guides/Permits-for-Higher-Education-Institutions is the page most likely to answer this and it failed TLS verification ('unable to verify the first certificate') on two attempts — the site is live, not dead, and should be retried from a client with a working cert chain. Circumstantial and NOT scored: international branch campuses demonstrably operate in Dubai free zones under dual MoHESR/KHDA licences (per Gulf News reporting), which strongly implies foreign entities can be licensed — but implication from the existence of licensees is not the licensing rule, and this gate is the school's own right to exist, so I will not infer it. Resolving this needs the KHDA/CAA licensure standards document or qualified legal review. SOURCE QUOTE, verbatim from the page on the retrieval date: “The commission licenses institutions and accredits their academic programmes in line with international standards.” | seed | seed | 9f105254-aacc-4913-b664-48ad346ed60a | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 18 | 5bfdf345-180e-4bb7-9a10-a6a2259c0377 | ae | tax-burden | UNRESOLVED | USD/member-month | https://u.ae/en/information-and-services/finance-and-investment/taxation/corporate-tax | The Official Platform of the UAE Government (u.ae) | 2023-06-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed 136 and an independent agent re-fetched the source and rejected it. VERIFIER: The quote '0 per cent for taxable income up to AED 375,000' is literally present, but the page states only the corporate tax rates (0% and 9%). The number 136 does not appear on the page and cannot be derived from the quote. The quote is real but does not establish the claimed value. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: TREAT THE RATES AS SOURCED AND THE NUMBER AS MODELLED — the base is my assumption, not a fact. RATES (both sourced, both OFFICIAL_PRIMARY): corporate tax 0% up to AED 375,000 and 9% above, per the quote below (u.ae, Federal Decree-Law No. 47 of 2022, applies to financial years beginning on or after 1 June 2023); VAT 5%, per u.ae 'The rate of VAT is 5 per cent.' (https://u.ae/en/information-and-services/finance-and-investment/taxation/vat/valueaddedtaxvat, introduced 1 January 2018). BASE (ASSUMPTION, NOT SOURCED): programme fee of USD 2,000 per member-month. I did not source Network School's actual fee; if the engine holds the real fee, recompute — the result scales almost linearly with it. CONVERSION: VAT = 5% x 2,000 = USD 100/member-month. Corporate tax = assumed 20% operating margin x 2,000 = USD 400 profit/member-month x 9% = USD 36/member-month. Total = 100 + 36 = USD 136/member-month. ASSUMPTIONS THAT COULD MOVE THIS A LOT: (1) I ignored the AED 375,000 (~USD 102,000) 0%-band, which is immaterial per-member at scale but would zero the CT component for a small cohort; (2) I assumed a MAINLAND entity — a Qualifying Free Zone Person earning Qualifying Income pays 0% CT, which would drop this to USD 100/member-month (VAT only), and the QFZP qualifying-activity list was narrowed by Ministerial Decision No. 229 of 2025, which I did not read; (3) I assumed education/training supplies are standard-rated for VAT — the UAE zero-rates some education supplies, and if that applied the figure could fall toward USD 36 or lower. I did not source a 0% personal-income-tax statement from an official page, so no PIT component is included and none is claimed. The 20% margin is my assumption and is not sourced. QUOTE THE RESEARCHER OFFERED: “0 per cent for taxable income up to AED 375,000” | seed | seed | 3625f070-4d9f-475c-9a38-375f35e87ad9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 19 | fdc1a764-4593-4a72-9fa8-7a266c8d0066 | ae | import-duty | UNRESOLVED | USD | https://www.sharjahcustoms.gov.ae/en/customs-tariff | Sharjah Customs (Government of Sharjah, UAE) | 2003-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed 7500 and an independent agent re-fetched the source and rejected it. VERIFIER: The quote about 5% duty on imported goods is literally present, but the figure 7500 appears nowhere on the page. 7500 would be 5% of an assumed 150,000 base, but that base comes from the researcher, not the source. The quote establishes a rate, not the claimed absolute value. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: RATE SOURCED, BASE ASSUMED. RATE: 5% ad valorem on CIF value, per the quote below (Sharjah Customs — an emirate customs authority applying the GCC Common Customs Tariff, in force since 1 January 2003). BASE (ASSUMPTION, NOT SOURCED): USD 150,000 CIF for a one-off campus fit-out of AV/IT/networking/office equipment. CONVERSION: 5% x 150,000 = USD 7,500 one-off. TARIFF LINE — READ THIS: I did NOT source a specific HS line. I applied the GENERAL 5% rate because I could not verify the line-level rate for the relevant headings (e.g. HS 8471 automatic data-processing machines, HS 8517 communications apparatus), some of which may be duty-free or reduced under the GCC tariff schedule. If any of those lines are 0%, this figure is an OVERSTATEMENT, possibly a large one. MATERIAL STRUCTURAL POINT: goods imported into a UAE free zone are duty-suspended and duty falls due only on entry to the mainland — a free-zone-sited campus could face USD 0 of this. The ICP customs page (https://icp.gov.ae/en/uae-customs-en/customs-union-for-gcc-states/) returned 404 on direct fetch; Sharjah Customs was used as the primary substitute and is itself an official UAE customs authority, not a downgrade in tier. QUOTE THE RESEARCHER OFFERED: “5% duty is imposed on goods imported from abroad excluding goods in transit to GCC countries.” | seed | seed | 36397be5-1093-47d8-b5b9-cd34ca0adc29 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 20 | 5f5cc2fe-8510-42a6-9192-507988960d61 | ae | political-stability | 79.37 | WGI 0-100 score | https://data360api.worldbank.org/data360/data?DATABASE_ID=WB_WGI&REF_AREA=ARE&INDICATOR=GOV_WGI_PV | World Bank — Worldwide Governance Indicators (via Data360 API) | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | ANALYST_VERIFIED | OFFICIAL_API | NULL | UAE Political Stability and Absence of Violence/Terrorism, 2024 (latest available), score 79.366146 on the 0-100 scale. DIRECTION: higher is better, as the criterion requires. METHODOLOGY CAVEAT THE ENGINE SHOULD CARRY: the criterion asks for a 'WGI percentile 0-100', but WGI's 2025 methodology revision REPLACED the percentile rank with a 0-100 score anchored to fixed reference points. The legacy percentile-rank indicators (PV.PER.RNK, RL.PER.RNK) are retired from the main World Bank API and now return 'The indicator was not found. It may have been deleted or archived.' I therefore used the WGI_SC breakdown (the 0-100 anchored score), which is the closest available successor and is on the same 0-100 range and direction, but it is NOT numerically identical to a percentile rank and should not be compared against percentile values captured before the revision. 90% confidence interval: 72.66 to 86.08 (WGI_SC_LB / WGI_SC_UB). 2023 value was 79.268472, so the series is flat. SOURCE QUOTE, verbatim from the page on the retrieval date: “"OBS_VALUE": "79.366146", ... "INDICATOR": "GOV_WGI_PV", "REF_AREA": "ARE", ... "COMP_BREAKDOWN_1": "WGI_SC", ... "TIME_PERIOD": "2024"” | seed | seed | b805e81f-ba24-469d-bddf-32e5aa6c273c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 21 | 4e754728-0114-4b45-a872-a12ba3a16eb1 | ae | rule-of-law | 65.52 | WGI 0-100 score | https://data360api.worldbank.org/data360/data?DATABASE_ID=WB_WGI&REF_AREA=ARE&INDICATOR=GOV_WGI_RL | World Bank — Worldwide Governance Indicators (via Data360 API) | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | ANALYST_VERIFIED | OFFICIAL_API | NULL | UAE Rule of Law, 2024 (latest available), score 65.515759 on the 0-100 scale. Higher is better. Same methodology caveat as political-stability: this is the WGI_SC 0-100 anchored score introduced by the 2025 WGI methodology revision, NOT the retired percentile rank (RL.PER.RNK now returns 'The indicator was not found' from the main World Bank API). Same range and direction, not numerically interchangeable with pre-revision percentiles. 90% confidence interval: 61.26 to 69.77. 2023 value was 64.697283 — a mild upward drift. Note this sits ~14 points below political-stability, which is the characteristic UAE pattern: very stable, less strong on rule-of-law measures. SOURCE QUOTE, verbatim from the page on the retrieval date: “"OBS_VALUE": "65.515759", ... "INDICATOR": "GOV_WGI_RL", "REF_AREA": "ARE", ... "COMP_BREAKDOWN_1": "WGI_SC", ... "TIME_PERIOD": "2024"” | seed | seed | c45e1e42-a744-4279-9c94-fb1b0c0e41b1 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 22 | bf31b152-ee2f-4640-b37f-0d9b6d4f4d8f | ae | currency-inflation | 1.3 | CPI %/yr | https://www.imf.org/external/datamapper/api/v1/PCPIPCH/ARE | International Monetary Fund — World Economic Outlook (DataMapper API) | 2025-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | ANALYST_VERIFIED | OFFICIAL_API | NULL | IMF WEO 'Inflation rate, average consumer prices' (PCPIPCH) for ARE. 2025 = 1.3%/yr, used as the value (most recent reference year at or before retrieval). Series context: 2022 = 4.8, 2023 = 1.6, 2024 = 1.7, 2025 = 1.3, 2026 (WEO projection) = 2.5. Caveat: the 2025 figure is a WEO estimate rather than a confirmed outturn, and the 2026 projection of 2.5 is a forecast and is NOT the value. Low and stable, consistent with the dirham's peg to the US dollar — UAE inflation is largely imported from US monetary policy plus local housing costs. SOURCE QUOTE, verbatim from the page on the retrieval date: “"ARE":{... "2023":1.6,"2024":1.7,"2025":1.3,"2026":2.5 ...}” | seed | seed | 8e87cf53-686c-4380-9c59-8bccca1feda4 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 23 | 2b9493a2-10d5-4da4-9d53-2630ec207855 | ae | corruption | 69 | TI CPI 0-100 | https://www.transparency.org/en/countries/united-arab-emirates | Transparency International | 2025-12-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | MODEL_ASSISTED_RESEARCH | NULL | UAE CPI score 69/100 (higher is better), Transparency International 2025 Corruption Perceptions Index. Flagged UNVERIFIED rather than ANALYST_VERIFIED for one specific reason: the fetched page reported the rank as '21 out of 182 countries', but the CPI has historically covered 180 countries, so either the coverage expanded in the 2025 edition or the fetcher garbled the denominator. The SCORE of 69 is the value this criterion needs and I have no reason to doubt it; the rank denominator is the part I could not confirm and it does not affect the value. A re-fetch should confirm both. Licensing: TI CPI is explicitly permitted for this product. SOURCE QUOTE, verbatim from the page on the retrieval date: “a score of 69 this year” | seed | seed | a20ed28b-00fc-4280-beb3-98f11b9425cd | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 24 | ee68a443-1bef-4bd7-9810-8b02aad2c08a | ae | english-proficiency | 487 | EF EPI score | https://www.ef.com/wwen/epi/regions/middle-east/united-arab-emirates/ | EF Education First | 2025-01-01 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | MODEL_ASSISTED_RESEARCH | NULL | EF EPI score 487, global rank #72, placing the UAE in EF's 'Moderate Proficiency' band. Two caveats. (1) EDITION YEAR NOT CONFIRMED: the country page states the score and rank but does not state the edition year anywhere the fetcher could see. I set effectiveAt to 2025-01-01 on the assumption that a page fetched in July 2026 carries the most recent edition (published late 2025), but this is an ASSUMPTION and the date should not be relied on — the score is what is sourced, the year is not. (2) LICENSING UNRESOLVED, as the brief notes: EF is the only source for this metric and its commercial-use terms have not been cleared for this product. (3) INTERPRETIVE NOTE the engine should carry: EF EPI measures proficiency among people who take EF's free online test — for the UAE this is a poor proxy for the operating environment, because English is the de facto working language of business in Dubai/Abu Dhabi while the tested population skews toward non-English-dominant residents. The score likely understates practical English availability for a Network School cohort by a wide margin. SOURCE QUOTE, verbatim from the page on the retrieval date: “EF EPI score: 487” | seed | seed | 7c1cc92e-cf4e-4019-8d54-5a4bd2314589 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 25 | fd6cebeb-ae67-40de-afcf-2252cdf9aae0 | ae | entry-friction | UNRESOLVED | days | https://www.gdrfad.gov.ae/en/services/64154a31-ec6d-11ec-140b-0050569629e8 | General Directorate of Residency and Foreigners Affairs — Dubai (GDRFA) | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed 2 and an independent agent re-fetched the source and rejected it. VERIFIER: The quote about a one-year extendable period of residence is literally present in the Terms and Conditions section. However, it speaks only to visa duration and says nothing about entry friction, number of steps, or processing burden. The claimed value of 2 is an analyst score on an unstated scale that the quote does not establish. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: 2 days = the 48-hour completion time GDRFA states for the Virtual Work Visa. Cohort-weighted value equals the raw value because no nationality differentiation is stated on the GDRFA or u.ae pages for this route — so the US/EU/IN/CN weighting (30/25/25/20) collapses to a single figure. SCOPE OF WHAT THIS NUMBER COVERS — important: 48 hours is GDRFA's stated PROCESSING time for the visa application itself. It EXCLUDES steps that a member actually has to complete before they are resident: the medical fitness test, health insurance procurement, Emirates ID issuance, and document assembly (as of 27 January 2026, six months of bank statements showing regular salary deposits are reportedly required — I did not verify that from an official page and do not score it). Real door-to-door time is therefore meaningfully longer than 2 days, but I have not sourced a figure for the full journey and will not invent one. Marked UNVERIFIED because the sourced number and the metric's intent are not the same thing: if the engine means 'days to obtain the visa' literally, 2 is right; if it means 'days until lawfully resident and operational', 2 is an understatement. QUOTE THE RESEARCHER OFFERED: “The virtual work has a one-year, extendable period of residence from the date of issuance.” | seed | seed | 34fa0969-1085-423c-836d-a352dda38d81 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 26 | 874d2e94-7310-4e71-be33-5ac449e66313 | ae | expropriation-risk | UNRESOLVED | property-rights index 0-100 | https://www.heritage.org/index/pages/country-pages/unitedarabemirates | The Heritage Foundation — Index of Economic Freedom | 2026-07-17 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | UNRESOLVED on a fetch failure, not on substance. Heritage's Index of Economic Freedom Property Rights component is the intended source (Heritage and Fraser are both permitted; ICSID and UNCTAD are correctly excluded as non-commercial and I did not touch them). BLOCKED: the country page https://www.heritage.org/index/pages/country-pages/unitedarabemirates 301-redirects to https://economicfreedom.heritage.org/, and that host returned HTTP 403 to the fetcher. The index is LIVE — this is a bot block, not a dead source. I deliberately did NOT substitute a worse source (relocation sites and aggregators reproduce Heritage numbers without provenance, and quoting them would launder an unverified figure through a real-looking citation). DIRECTION REMINDER for whoever resolves this: this is a RIGHTS score — HIGHER IS BETTER — so a strong UAE property-rights reading should raise, not lower, this criterion. RELATED SOURCED EVIDENCE that bears on the underlying question but is NOT this metric: Dubai Law No. (7) of 2006 Art. 4 gives foreign holders statutory freehold/99-year leasehold rights in designated areas (see foreign-title-statute above), and WGI Rule of Law 2024 = 65.52/100. Neither is a property-rights index and neither should be substituted for one. RESOLUTION PATH: fetch economicfreedom.heritage.org from a browser-class client, or use Fraser's Economic Freedom of the World 'Legal System & Property Rights' area as the permitted alternative. | seed | seed | 8a5d3df6-1938-49c7-a43e-e1a85d9b179b | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 27 | 5f6e28e0-7c50-4772-a328-cbeaac0d4651 | ae | government-receptivity | 3 | rubric 0-5 | https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae | The Official Portal of the UAE Government (u.ae) | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 3: Named policy fit: a programme this project class demonstrably qualifies for, published eligibility criteria (SEZ regime, education-hub licence, digital-nomad statute). THE FACTS: · The UAE operates a published federal virtual-work (remote-work) residence route with stated eligibility criteria — anchor 3's own named example, a digital-nomad statute. The Network School node's cohort is remote workers and founders employed by entities outside the UAE, which is exactly the population the route is written for. https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae “The one-year visa allows them to enter the UAE under self-sponsorship and work in line with the terms and conditions issued with the visa.” · Eligibility is criteria-based and published: a stated income floor and proof of employment outside the UAE. This is published eligibility criteria, not a discretionary welcome. https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae “salary certificate of a minimum of 3,500 (three thousand five hundred) US dollars or equivalent in foreign currency” · The UAE has a named, in-force licensing regime for the virtual-asset activity the project class also involves, published by the Dubai regulator — corroborating that named policy fit exists across more than one leg of the project class, not just the visa. https://www.vara.ae/en/ “The Virtual Assets Regulatory Authority (VARA) is responsible for regulating and overseeing the provision, use, and exchange of virtual assets in and from the emirate of Dubai.” FACTS: The UAE publishes a federal virtual-work residence route on the government's own portal with objective, published eligibility criteria (one-year self-sponsored permit; USD 3,500/month salary certificate; proof of employment outside the UAE; health insurance). The Network School node's cohort — remote workers and founders paid by non-UAE entities — demonstrably qualifies. Dubai separately operates VARA, a named in-force licensing regime covering the project's virtual-asset leg. MAPPING: anchor 3. Anchor 3 names 'digital-nomad statute' as a qualifying instrument and requires the class to demonstrably qualify against published criteria; both hold. It is not anchor 2 ('generic openness only') because this is not an FDI welcome — it is an activity-specific statute whose criteria the class meets. WHAT ARGUED FOR HIGHER AND WHY IT DID NOT REACH: Anchor 4 requires a named empowered official or agency to have met, corresponded, or issued a public statement about THIS SPECIFIC PROJECT or a close analogue, ON THE RECORD. Secondary write-ups assert that Abu Dhabi and Dubai government representatives engaged with the Network School and that a Dubai campus is planned, and an ADGM speaker appeared at the Network State Conference 2025 — but every one of those traces is a personal blog, an SEO aggregator, or a press summary, i.e. exactly the source class the packet forbids. I found no on-the-record statement by a named UAE official or agency about the Network School or a close analogue, and I did not fetch one. Absent that instrument, anchor 4 does not reach however plausible the story, and anchor 5 (signed MOU / gazetted incentive / allocated land or licence naming the project or its class) has no instrument at all. WHAT WOULD MOVE IT: a WAM or mediaoffice.ae release quoting a named ICP/DET/ADGM/Executive Council official on the Network School or an equivalent residential-school-for-remote-workers node moves it to 4; a signed MOU or allocated licence naming the class moves it to 5. Documented official hostility toward the class — a probe, a revocation, a named official opposing it — would drop it toward 0; none found. BLOCKS: moet.gov.ae's Dubai International Academic City page (the education-hub-licence leg, which would have independently supported anchor 3) resolved to a login wall and returned no substantive content; that leg is therefore excluded from this score rather than asserted. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: All three quotes re-fetched and literally present. u.ae carries 'The one-year visa allows them to enter the UAE under self-sponsorship and work in line with the terms and conditions issued' and 'salary certificate of a minimum of 3,500 (three thousand five hundred) US dollars or equivalent in foreign currency'; vara.ae carries the VARA remit sentence verbatim. MAPPING HOLDS: anchor 3 names 'digital-nomad statute' as a qualifying instrument and requires the class to demonstrably qualify against published criteria. The virtual-work visa is that statute, its criteria are published and objective (USD 3,500 salary certificate, proof of employment outside the UAE, health insurance, medical fitness), and the cohort of remote workers paid by non-UAE entities is the population the route is written for. Not anchor 2: this is not a generic FDI welcome but an activity-specific route. The note correctly refuses anchor 4 for want of any on-the-record statement by a named UAE official about this project or a close analogue, and correctly refuses anchor 5 for want of any instrument. It also declares the moet.gov.ae block rather than asserting the education-hub leg. This is the one claim that resists the upward pull. | seed | seed | 39841ae1-89cc-44c9-9a2b-a93c1030bdd3 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 28 | 73c799ba-620f-4072-a498-8efe3c029b3f | ae | visa-preference | 3 | rubric 0-5 | https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae | The Official Portal of the UAE Government (u.ae) | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | UNVERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A preferential route exists, is in force, open AS OF RIGHT on published criteria, covers ALL FOUR cohorts. THE FACTS: · A preferential route beyond tourist/standard business entry exists and is in force: a one-year self-sponsored residence permit for people working for a company outside the UAE. https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae “The one-year visa allows them to enter the UAE under self-sponsorship and work in line with the terms and conditions issued with the visa.” · The route's activity scope is residence plus remote work for a non-UAE employer — the project cohort's actual activity. https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae “A foreigner may live in the UAE while working for a company outside the UAE.” · Published criteria are objective and financial/documentary, and carry no nationality condition — no exclusion or quota touching the US/EU/IN/CN cohorts appears on the published criteria. https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae “salary certificate of a minimum of 3,500 (three thousand five hundred) US dollars or equivalent in foreign currency” REMAPPED BY ADVERSARIAL VERIFICATION (D-032/D-035). The scorer claimed 4 at "anchor 4: A preferential route exists, is in force, open AS OF RIGHT on published criteria, covers ALL FOUR cohorts.". An independent agent re-checked the facts against the pages AND the mapping against contracts/rubric.md, and the facts do not reach that anchor. VERIFIER: Facts hold: all three quotes are verbatim on u.ae ('A foreigner may live in the UAE while working for a company outside the UAE' appears as 'Under a virtual work visa, a foreigner may live in the UAE while working for a company outside the UAE' — substance intact). MAPPING FAILS — this is inferring upward from silence. Anchor 4 imposes two AFFIRMATIVE requirements: 'open AS OF RIGHT on published criteria' and 'covers ALL FOUR cohorts'. On re-fetch the page states no nationality condition, no quota, no pilot status, AND no statement that the permit is granted as of right — it is simply silent on all of it, and the page names ICP and GDRFA Dubai as the issuing authorities, i.e. approval bodies. The note's own reasoning inverts the burden: it scores 4 because anchor 3's disqualifiers are 'not affirmatively evidenced'. But anchors are ceilings, not defaults. Silence about discretion is not evidence of entitlement, and silence about nationality is not evidence that all four cohorts are covered — it is absence of the instrument that anchor 4 requires, which caps the score. The note itself concedes 'as of right is the weakest link' and that 'u.ae nowhere states the permit is granted as of right'. That concession is the answer. Anchor 3 — 'a preferential route exists and fits, but is discretionary' — is what the sourced facts support: a real, in-force, fitting route whose grant sits with ICP/GDRFA and is nowhere published as an entitlement. Score 3. (Anchor 5 is correctly excluded: the published term is one year, not multi-year.) ORIGINAL SCORING NOTE, retained — its facts stand and only the mapping moved: FACTS: The UAE virtual-work residence visa is in force, self-sponsored, one year, and open on published criteria (USD 3,500/month salary certificate, proof of employment outside the UAE, health insurance, medical fitness test); applications go to ICP or GDRFA Dubai. The published criteria state no nationality condition, no quota, and no pilot status, so no cohort among US/EU/IN/CN is excluded on the face of the instrument. MAPPING: anchor 4. WHAT ARGUED FOR HIGHER AND WHY IT DID NOT REACH: anchor 5 is genuinely tempting — this IS a remote-work route and the project squarely fits it, which is 5's first limb. It fails 5's second limb outright: 5 requires 'published multi-year terms', and the published term is one year. The Golden Visa offers ten years but is talent/investor/student-gated on recommendation and capital thresholds, so it is not a route this cohort holds as of right and cannot be substituted in (and the gate forbids combining two routes). So the ceiling is 4. WHAT ARGUED FOR LOWER: two things. (1) 'As of right' is the weakest link — u.ae publishes objective criteria but nowhere states the permit is granted as of right, and UAE residence approvals sit with ICP/GDRFA. I score 4 rather than 3 because anchor 3's disqualifiers are specific — discretionary, capped, pilot-stage, or not in force — and none of the four is affirmatively evidenced; a sourced statement that ICP retains approval discretion, or evidence of a cap, drops this to 3. (2) Anchor 2's 'wrong activity' argument: the route permits remote work for a foreign employer and residence, but says nothing about enrolment in a UAE educational institution. It does not hold, because the node's cohort activity is remote work for non-UAE entities and residence — not accredited enrolment. If the node's model requires members to study at a UAE-licensed institution, this route is the wrong activity and the score is 2. WHAT WOULD MOVE IT UP: a published multi-year remote-work or education route the class fits as of right takes it to 5. A named restriction, quota or exclusion touching US/EU/IN/CN would take it to 0; none found. VERIFICATION LIMIT: single-source (u.ae). Secondary sources assert the route is open 'regardless of your nationality' and that a six-month bank-statement requirement was added, but those are relocation-firm and nomad-blog pages the packet forbids, so I did not rely on them — hence UNVERIFIED rather than CORROBORATED. | seed | seed | 8cd0b7a8-165d-4b03-92a8-90eca15c867a | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 29 | 0cc9f7f7-26f8-406c-8a7e-0fe52b7988fa | ae | web3-support | UNRESOLVED | NULL | https://mediaoffice.ae/en/news/2025/may/12-05/dubai-finance-signs-mou-with-cryptocom | Dubai Media Office / Government of Dubai | 2025-05-12 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | ANALYST_ENTRY | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The scorer claimed 5 and the facts did not survive re-fetch. VERIFIER: The score fails on the exact leg the rubric says it cannot skip. The rubric's own NOTE: 'anchors 1 and 3 both turn on BANKING, which DATA-SOURCES.md §194 says has no automatable source. A score that skips banking is not sourced.' Both banking facts fail. (1) The CBUAE Payment Token Services Regulation URL returned HTTP 403 to WebFetch — I could not check it; that is a block, not a bad source. (2) The Dubai Finance banking claim is CONTRADICTED by its own page. The quote is near-verbatim ('securely convert these payments into Emirati dirhams and transfer them to Dubai Finance accounts'), but the claim built on it — 'a regulated crypto-to-bank path is in LIVE USE by the government itself' — is false: the same release says 'Once the necessary technical arrangements for the agreement's activation are finalised, individuals and businesses ... will be able to pay service fees'. It is an unactivated MoU. Under the contract's own warning, an intention is not a commitment; a signed MoU pending technical activation is not acceptance in operation, so it does not carry anchor 5's 'acceptance of digital assets' either. The other facts DO hold and are strong — the Digital Dirham transaction was genuinely executed on 11/11/2025 with named officials, and the VARA register carries the IPA-prohibition quote verbatim with Binance FZE, OKX and Deribit holding full licences (minor drift: I counted 9-11 Exchange Services licensees against the claimed 13, and could not confirm Foris DAX; immaterial to the leg). But with one banking source 403 and the other contradicted, rung 3 is unsourced, and the rubric says the ladder is not sourced without it. Degrade rather than remap: the failure is evidentiary, not a mis-mapping. ORIGINAL SCORING NOTE: FACTS: (a) Dubai Finance signed an MoU on 12 May 2025 to enable payment of government service fees using cryptocurrencies, signed by Ahmad Ali Meftah (Executive Director, Central Accounts Sector, Dubai Finance) in the presence of H.E. Abdulla Mohammed Al Basti (Secretary General, Executive Council of Dubai) and H.E. Abdulrahman Saleh Al Saleh (Director General, Dubai Finance). (b) On 11 November 2025 the Ministry of Finance, Dubai Finance and the CBUAE executed the UAE's first government transaction using the Digital Dirham. (c) VARA's public register shows 51 entities and 13 Exchange Services licensees (Binance FZE, OKX, Deribit, Crypto.com's Foris DAX ME FZE), with IPA holders barred from operating — so licensed exchanges demonstrably operate. (d) Banking, per the rubric's §194 note: the Dubai Finance flow converts crypto to dirhams for transfer into Dubai Finance's bank accounts, and CBUAE's Payment Token Services Regulation C 2/2024 (effective 31/8/2024, In-Force) makes payment-token issuance/conversion/custody a Central Bank-licensed financial activity. MAPPING: anchor 5 — 'acceptance of digital assets' by a named government programme, which is precisely what the Dubai Finance instrument and the Digital Dirham programme are. Rungs 3 and 4 are also cleared: framework plus operating licensed exchanges plus obtainable regulated fiat rails (3), and published federal tax treatment plus documented public-sector engagement (4). WHAT ARGUED FOR LOWER AND WHY IT DID NOT WIN: the §194 banking requirement is the real pressure point, and I do not want to overclaim it. What I sourced is institutional banking access — a government department banking crypto-derived dirham receipts, and a Central Bank licensing regime for token conversion. What I did NOT source, and could not, is the retail/operational fact the rubric is really pointing at: whether an ordinary VARA-licensed or free-zone entity can open and keep a normal UAE operating bank account without de-risking. There is no primary source for that (which is what §194 says). If that were shown to fail in practice, the ladder's rung 3 breaks and this collapses toward 1 regardless of the public-sector adoption. WHAT WOULD MOVE IT DOWN: a documented pattern of UAE banks declining service to licensed VASPs takes it to 1; withdrawal of the government fee-payment programme takes it to 4 or below. It cannot go higher — 5 is the ceiling. NOTE ON SOURCE TIER: mediaoffice.ae is the Government of Dubai's own announcement channel but is not Dubai Finance itself, hence OFFICIAL_SECONDARY; the VARA register and the CBUAE rulebook are OFFICIAL_PRIMARY and corroborate the framework and rails legs independently. | seed | seed | 6e4fe8f4-e981-4acf-9831-3e356acea696 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 30 | 0954ab8f-9012-48ee-b590-b6f8fc8fdd53 | ae | capital-controls | UNRESOLVED | NULL | https://www.elibrary.imf.org/view/journals/002/2025/327/article-A002-en.xml | International Monetary Fund — UAE Staff Report for the 2025 Article IV Consultation, Informational Annex (IMF Country Report, 11 November 2025) | 2025-11-11 | 2026-07-17T00:00:00.000Z | MULTILATERAL | NEEDS_EVIDENCE | ANALYST_ENTRY | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The scorer claimed 1 and the facts did not survive re-fetch. VERIFIER: The two load-bearing facts could not be checked. https://www.elibrary.imf.org/view/journals/002/2025/327/article-A002-en.xml returned HTTP 403 Forbidden to WebFetch — so both the Article VIII / Decision 144 quote and the AED 3.6725 peg quote are unverified. I am reporting a 403, not impugning the IMF. The two u.ae facts DO hold verbatim ('100 per cent profit repatriation'; 'low foreign exchange controls'; 'strong appreciation and fully convertible currency'). But the government's own investment-marketing page cannot carry an anchor on its own — it is precisely the promotional posture the contract says not to read as fact, and it resolves nothing about the capital account. This leaves the score resting entirely on a source I could not fetch, compounding a declared block: the note already concedes AREAER — the source the rubric NAMES for this subject — is login-gated and that 'the capital-account rows ... are therefore NOT verified'. The rubric's source discipline requires that substitution be declared, and the note does declare it honourably; but with the substitute ALSO 403, there is no verified evidence left. I note the analytic reasoning was sound and admirably refused 0 on the government's own 'low, not none' hedge — the problem is evidentiary, not directional. Anchor 1 may well be right; it is currently unsourced. ORIGINAL SCORING NOTE: SOURCE DISCIPLINE DECLARATION (required by the rubric): the rubric names IMF AREAER specifically. I could not use it. AREAER Online (https://www.elibrary-areaer.imf.org/) 302-redirects to an IMF Bookstore auth gateway and is login-gated; the IMF's own PDF host (imf.org/-/media/...) returned HTTP 403 to both WebFetch and curl. I report the block rather than pretend to the source. The capital-account rows — which is exactly where AREAER would resolve this subject — are therefore NOT verified. What I substituted is the IMF's own 2025 Article IV informational annex (same institution, MULTILATERAL, dated 11 November 2025), plus the UAE government's own investment portal. FACTS: exchange system free of restrictions on CURRENT international transactions except national-security restrictions notified under Decision 144; conventional peg at AED 3.6725/USD, de jure and de facto identical; official portal claims 100% profit repatriation and a fully convertible currency, and characterises the regime as having 'low foreign exchange controls'. MAPPING: anchor 1 — no approval requirement for transfers is documented anywhere I fetched, and no delays are documented, but the regime is not control-free: AML/CFT reporting obligations bite on transfers and Decision 144 security restrictions are on the record. That is 'reporting or registration only; no approval needed; no documented delays'. WHAT ARGUED FOR LOWER (0) AND WHY IT DID NOT REACH: the case for 0 is strong on its face — the IMF confirms current-transaction freedom, the peg is real, and the state promises 100% profit repatriation. It does not reach because anchor 0 requires 'No controls', full stop, and free repatriation of CAPITAL as well as income. The IMF sentence covers current transactions only and is not evidence about the capital account. The UAE's own portal says 'low' controls, not none — I will not read the government's own hedge as a zero. And the one source that would resolve the capital account, AREAER, is blocked. Scoring 0 here would be exactly the unstated-analyst-judgment the previous pass was rejected for. WHAT WOULD MOVE IT: an AREAER capital-account row showing no controls on repatriation of capital moves this to 0; a documented approval requirement for large transfers moves it to 2; documented discretionary approval or exit taxation moves it to 3. NOTE ON THE HARD CAP: this is nowhere near 5, so the Risk-axis cap is not triggered. | seed | seed | 761529e8-c592-4963-838c-aff71e761156 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 31 | 8bbacd2b-159e-4ae8-b49d-f4fd39daf103 | ae | policy-reversal-precedent | 2 | rubric 0-5 | https://tax.gov.ae/Datafolder/Files/Guides/CT/Free%20Zone%20Persons%20-%2020%2005%202024%20final%20for%20GCD.pdf | Federal Tax Authority (UAE) — Corporate Tax Guide: Free Zone Persons, CTGFZP1, May 2024 | 2023-06-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 3: Terms tightened AND applied to existing holders, with a transition period. THE FACTS: · THE INSTANCE: UAE free zones had granted tax incentives ('tax incentive periods', up to 50 years) under their own legislation. Federal Decree-Law No. 47 of 2022 imposed a federal corporate tax over the top of them, applying to tax periods commencing on or after 1 June 2023 — i.e. applied to free zone entities that already held those incentives, roughly eight months after the law was published in the Official Gazette. That gazette-to-effect gap is the transition period. https://tax.gov.ae/Datafolder/Files/Guides/CT/Free%20Zone%20Persons%20-%2020%2005%202024%20final%20for%20GCD.pdf “of 2022 on the Taxation of Corporations and Businesses ("Corporate Tax Law") was issued on 3 October 2022 and was published in Issue #737 of the Official Gazette of the United Arab Emirates ("UAE") on 10 October 2022. The Corporate Tax Law provides the legislative basis for imposing a federal tax on corporations and Business profits ("Corporate Tax") in the UAE. The provisions of the Corporate Tax” · THE GRANDFATHERING THAT DID SURVIVE: the incentive PERIOD an existing free zone holder was granted is expressly preserved — the 0% rate runs for the remainder of it. This is the fact that argues the reversal down to anchor 2. https://tax.gov.ae/Datafolder/Files/Guides/CT/Free%20Zone%20Persons%20-%2020%2005%202024%20final%20for%20GCD.pdf “The 0% Corporate Tax rate applies for the remainder of the tax incentive period stipulated in the legislation of the Free Zone in which the QFZP is registered.” · THE TIGHTENING THAT WAS APPLIED TO EXISTING HOLDERS: the preserved 0% is now conditional and partial. An existing free zone entity keeps 0% only if it is a Qualifying Free Zone Person and only on Qualifying Income; enumerated income (foreign or domestic permanent establishment income, most immovable-property income) is taxed at 9% notwithstanding the incentive it previously held. Existing holders were not left on their old unconditional terms. https://tax.gov.ae/Datafolder/Files/Guides/CT/Free%20Zone%20Persons%20-%2020%2005%202024%20final%20for%20GCD.pdf “income from the following sources (even if within the list above) will not give rise to Qualifying Income and will be taken into account in determining the Taxable Income that is subject to the 9% Corporate Tax rate” REMAPPED BY ADVERSARIAL VERIFICATION (D-032/D-035). The scorer claimed 3 at "anchor 3: Terms tightened AND applied to existing holders, with a transition period.". An independent agent re-checked the facts against the pages AND the mapping against contracts/rubric.md, and the facts do not reach that anchor. VERIFIER: Facts hold in the narrow sense: I pulled the FTA guide and confirmed all three quotes verbatim in the extracted text (the Official Gazette Issue #737 / 1 June 2023 passage; 'The 0% Corporate Tax rate applies for the remainder of the tax incentive period stipulated in the legislation of the Free Zone in which the QFZP is registered'; and the Qualifying Income / 9% exclusion passage). But the quotes do not establish the claim built on them. MAPPING FAILS. Anchor 3 requires 'terms tightened AND APPLIED TO EXISTING HOLDERS'. Two problems. (1) The 'up to 50 years' premise — the prior unconditional grant that is the whole predicate of a tightening — is NOT in the guide: grep over the full extracted text returns zero hits for '50 year', 'fifty', or 'grandfather'. The guide never documents what free zone holders previously held, so it cannot evidence that those terms were narrowed. That premise is analyst-supplied. (2) The ONLY thing the guide actually says about pre-existing incentive holders is that their incentive is PRESERVED — the 0% runs 'for the remainder of the tax incentive period'. That is grandfathering language, and the note's own reading ('what was preserved is the clock, not the terms') is an inference laid over a document that says the opposite of the anchor. What the sourced facts support is anchor 2: 'Terms tightened prospectively, existing holders grandfathered' — a new federal tax gazetted 10/10/2022, biting prospectively on tax periods from 1/6/2023, with the pre-existing incentive period expressly preserved. The note anticipates this exactly: 'If a verifier reads the CT law as a new tax rather than a modification of a granted incentive, this becomes 2.' I so read it, and the missing 50-year premise forces it. Score 2. (The instance is genuine and cited, so this is not the rubric's absence-of-history trap; 4 and 5 are correctly excluded.) ORIGINAL SCORING NOTE, retained — its facts stand and only the mapping moved: THE INSTANCE, AS THE RUBRIC DEMANDS: the UAE's free zones granted tax incentives of up to 50 years under their own legislation; Federal Decree-Law No. 47 of 2022 (gazetted 10 October 2022, applying to tax periods commencing on or after 1 June 2023) narrowed those incentives and applied the narrowing to entities that already held them. Per the FTA's own guide, an existing free zone entity now keeps 0% only if it qualifies as a Qualifying Free Zone Person and only on Qualifying Income; permanent-establishment and most immovable-property income is taxed at 9% regardless of the incentive previously granted. MAPPING: anchor 3 — 'terms tightened AND applied to existing holders, with a transition period'. All three limbs are present: tightened (0% became conditional and partial), applied to existing holders (no carve-out for pre-2023 entities), transition period (~8 months from gazette to effect, plus the incentive period itself preserved). WHAT ARGUED FOR A LOWER SCORE (2) AND WHY IT DID NOT WIN: the FTA expressly preserves 'the remainder of the tax incentive period', which reads like grandfathering, and one can fairly argue that a sovereign introducing a general corporate income tax where none existed is ordinary fiscal policy rather than a targeted revocation of a grant. I do not take it, because anchor 2 requires 'existing holders grandfathered' and existing holders were not: what was preserved is the clock, not the terms. An entity that held an unconditional exemption now pays 9% on income it was previously promised exemption from. If a verifier reads the CT law as a new tax rather than a modification of a granted incentive, this becomes 2. WHAT ARGUED FOR A HIGHER SCORE (4/5) AND WHY IT DID NOT REACH: nothing did. Anchor 4 requires a programme closed or suspended to new entrants with little notice, and anchor 5 requires grants revoked from existing holders or a regime abolished without transition. Neither exists here — the free zone regime remains open, the incentive survives, and there was notice. I searched for revocations of UAE visa or licence regimes granted to existing holders and found no instance I could source to anything but forbidden secondary content. WHY NOT 0 OR 1: anchor 1 requires no reversal found in the last ten years; this instance is a reversal-shaped event within that window, so 1 is unavailable. Anchor 0 additionally requires a documented history of honouring grandfathering, and the FTA guide documents the opposite on this instance. Note the rubric's own warning applies to the counterfactual, not to this: I am not scoring absence of history — I have a cited instance. VERIFICATION LIMIT: single-source (the FTA guide), hence UNVERIFIED. The FTA guide is OFFICIAL_PRIMARY and self-corroborating on all three limbs, but I did not independently fetch the Decree-Law text, Cabinet Decision No. 100 of 2023, or Ministerial Decision No. 265 of 2023 — the FTA guide's citations to them are the only evidence I have of their contents. A second fetch of the Decree-Law's free zone articles would take this to CORROBORATED. | seed | seed | 99e604f3-7240-44a1-8e49-faf5358e1b8a | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 32 | 597699f8-5c62-4053-8f11-b8914df67e3b | ae | visa-access | {"route":"Virtual Work Residence Visa (UAE remote-work / Virtual Working Programme)","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":365,"passportSet":"*"} | NULL | https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-working-outside-the-uae | The Official Platform of the UAE Government (u.ae) | 2021-03-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | D-037 RE-KEY, 2026-07-17. Four identical rows (US/EU/IN/CN) collapse into THIS one, keyed passportSet "*" — the route is stated without any nationality condition. The four rows carried one URL, one verbatim quote and one route between them; they existed only because D-036 keyed a claim by scenario cohort, and "EU" is not a bloc the u.ae page names. Recording four per-cohort claims asserted four things the page does not say. ANY_PASSPORT says what it does say. NOTHING about the evidence changed and the gate's verdict is unmoved: NEEDS_EVIDENCE, because the route bars STUDY and no route list was ever enumerated. The researcher's own caveat below is the reason "*" is honest here AND the reason it must be re-checked: nationality-neutrality is inferred from the absence of a restriction, not from an affirmative statement of its absence. D-036 RESHAPE, 2026-07-17. One route, one source. Nothing about the claim changed: the u.ae page establishes THIS route for THIS cohort, which is exactly what an adversarial verifier confirmed of the matrix on the same day. The matrix shape is gone because one sourceUrl cannot carry a claim drawn from a dozen pages; this one happened to come from a single page, which is why it survived verification when the other four countries' did not. NOT ENUMERATED: nobody has read the UAE's complete route list for this cohort, so a cohort with no qualifying route is NEEDS_EVIDENCE and not FAIL. That is D-036's point — absence of a qualifying route is indistinguishable from nobody having finished looking. NEAR-MISS — this route permits RESIDE and WORK_REMOTE but NOT STUDY, so it fails the all-three test for every cohort. Reasoning: the u.ae and GDRFA pages describe the visa purely as residence for remote employment with an entity outside the UAE; neither page grants any study right. Per the silence-is-not-permission rule I did not credit STUDY. maxStayDays=365: GDRFA says 'one-year, extendable', but extension requires a fresh application on re-proof of eligibility, i.e. it is not a renewal as of right, so I did not count it. Cohort-uniform: neither official page states any nationality restriction or nationality-differentiated condition, and the eligibility test is income-and-employment based (USD 3,500/month + health insurance + proof of work for a non-UAE employer), so I list the same route for US/EU/IN/CN. I have NOT verified nationality-neutrality against an affirmative government statement — the basis is the absence of any restriction on the two official pages. IMPORTANT OMISSIONS, deliberately not guessed: the UAE student visa (u.ae describes it as requiring sponsorship by an ACCREDITED university/college — Network School would not qualify, and it carries no remote-work right), the Green Visa (5yr, freelance/self-employment permit from MoHRE) and the Golden Visa (10yr) all exist, but I did not fetch a page with a verbatim quote establishing their permit sets in this pass, so they are omitted rather than invented. A later pass should check whether ANY UAE residence visa carries an express study right — if a residence-visa holder's right to enrol is stated anywhere official, the Golden/Green route could change this gate's outcome. Corroborated by GDRFA Dubai (https://www.gdrfad.gov.ae/en/services/64154a31-ec6d-11ec-140b-0050569629e8): 'The virtual work has a one-year, extendable period of residence from the date of issuance.' and 'a monthly income of no less than (3,500) US dollars'. SOURCE QUOTE, verbatim from the page on the retrieval date: “The one-year visa allows them to enter the UAE under self-sponsorship and work in line with the terms and conditions issued with the visa.” | seed | seed | e96a343e-4840-4dab-8941-25782bb33843 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 33 | 92384f05-b2a7-4271-8ccb-b38d1b7b227e | ae | visa-access | {"route":"Golden Visa (UAE long-term residence visa)","permits":["RESIDE","WORK","STUDY"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/golden-visa | The Official Platform of the UAE Government (u.ae), restating the Federal Authority for Identity, Citizenship, Customs & Port Security (page states 'Source: ICP') | 2026-03-24 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE, verbatim from the page on the retrieval date: “The UAE’s ‘Golden visa’ is a long-term residence visa which enables foreign talents to live, work or study in the UAE while enjoying exclusive benefits which include: a long-term, renewable residence visa valid for 5 or 10 years / the privilege of not needing a sponsor / the ability to stay outside the UAE for more than the usual period of six months needed to keep their residence visa valid / the ability to sponsor family members, including spouses and children.” THIS ANSWERS THE FIXTURE'S BLOCKED STUDY QUESTION: a UAE residence visa DOES carry an express STUDY right — the Golden Visa's own scope sentence is 'live, work or study in the UAE'. PERMITS: I recorded exactly the page's three verbs — RESIDE ('live'), WORK ('work'), STUDY ('study'). I did NOT record WORK_REMOTE: the page says 'work in the UAE' and nowhere states whether that includes remote work for a non-UAE employer; whether generic WORK subsumes WORK_REMOTE is a verdict question for the gate, not a research finding. Phrasing caveat: the verbs are joined by a disjunctive 'or' ('live, work or study'); I read the sentence as enumerating what the visa enables rather than as mutually exclusive alternatives, because the benefits list describes a single unconditioned residence permit — but the disjunction is on the page and a verifier should see it. maxStayDays=1825: the visa is a residence visa 'valid for 5 or 10 years' depending on category (10 years for public investors, exceptional talents, humanitarian pioneers; 5 years for real-estate investors, entrepreneurs, high-school achievers); 1825 is the conservative floor across categories, and for a residence permit the permitted stay runs the validity of the permit. RENEWAL NOT COUNTED: 'renewable' is stated but no page I fetched states renewal as of right, so I did not extend beyond one grant. ELIGIBILITY IS CATEGORY-GATED, NOT NATIONALITY-GATED: the route is open only to the page's enumerated categories (investors with AED 2M minimum capital, entrepreneurs, exceptional talents/PhD holders/specialists, outstanding students, humanitarian pioneers and frontline heroes) — a member must individually fit a category; that is an access condition the gate reader must weigh even though it is not a passport condition. PASSPORT SET '*': the page states no nationality condition anywhere — eligibility is talent/capital/achievement-based and the page's own subject is 'foreign talents'. BASIS IS THE WEAKER ONE: the page states none, and I did not find an affirmative nationality-neutrality statement (ICP was HTTP 500 sitewide on the retrieval date; the GDRFA Golden Residency page at https://www.gdrfad.gov.ae/en/services/335969f4-8045-11ed-4fe5-0050569629e8 renders only a sub-service index with no scope text). effectiveAt is the page's own 'Updated On: 24 Mar 2026' date — the decree date could not be sourced because the underlying instrument (Federal Decree-Law 29/2021 with Cabinet Resolution 65/2022) at uaelegislation.gov.ae/en/legislations/1601 returned HTTP 403 (Cloudflare) to this fetcher; that is a refusal of the fetcher, not evidence about the instrument. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK — HTTP 200 at the cited URL, independently fetched 2026-07-18. Page still shows 'Updated On: 24 Mar 2026' and 'Source: ICP'. FOLLOW-UP RESOLVED THE WORK_REMOTE QUESTION — AGAINST A PASS, 2026-07-18. A targeted follow-up chased the one question that could have flipped the UAE to the product's first visa PASS: does this route's 'work' cover remote work for a foreign employer, and are 'live/work/study' held simultaneously by one holder? Both resolve against it, so permits stay [RESIDE, WORK, STUDY] (WORK, not WORK_REMOTE) and the UAE stays NEEDS_EVIDENCE. Q1 — the u.ae Golden Visa page (re-fetched HTTP 200, 'Source: ICP', Updated 24 Mar 2026) grants the right to 'live, work or study IN THE UAE'; the work is in-UAE work, and the page never grants remote work for a non-UAE employer — that language belongs to the separate Virtual Work route. 'The privilege of not needing a sponsor' is residence self-sponsorship, not a remote-work authorization; MOHRE separately runs a 'Work Permits of Golden Visa Holders' service for UAE establishments employing holders 'from within the country' (WebSearch context — the MOHRE host ERR_CONNECTION_CLOSED on every attempt, so it is not a cited source). Per the do-not-stretch rule, WORK_REMOTE stays off. Q2 — the verbs are joined by a disjunctive 'or' with no primary text asserting all three run at once; not decisive on its own, but moot since the work is in-UAE regardless. No other single route qualifies (Virtual Work has no STUDY; Green no STUDY/no remote; Student no work). No verifier pass was run on this follow-up: it changed no gate value (permits are identical to the already-CORROBORATED claim) and added no pass-eligible claim — the deciding question resolved in the safe direction (no PASS), so there is no consequential new fact for an adversarial check to refute. UPGRADE PATH: read Cabinet Resolution 65/2022 and Federal Decree-Law 29/2021 primary text (both walled to the fetcher) to close the enumeration and confirm the in-UAE work scope against the statute. | seed | seed | e99d8354-8137-43af-a632-1b2195411de5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 34 | e044a3af-b277-4120-b880-f736d03de5db | ae | visa-access | {"route":"Green Visa (5-year self-sponsored residence visa)","permits":["RESIDE","WORK"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-doing-business-in-the-uae | The Official Platform of the UAE Government (u.ae), restating the Federal Authority for Identity, Citizenship, Customs & Port Security (page states 'Source: ICP') | 2026-02-26 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE, verbatim from the page on the retrieval date: “The Green visa is a type of long-term residence visa, which enables foreign talents to live, work or run a business in the UAE while enjoying exclusive benefits which include: a renewable residence visa valid for 5 years / the privilege of not needing a sponsor / the ability to sponsor family members, including spouses and children.” And: “The following categories of individuals are eligible for a Green visa: Skilled workers / Freelancers and self-employed / Investors and business partners.” URL CAVEAT A VERIFIER MUST KNOW: the u.ae index links this route as 'Green Visa' but the old slug .../residence-visas/green-visa returns 404; the live page renders at the residence-visa-for-doing-business-in-the-uae slug and its own breadcrumb and H1 read 'Types of visas > Green visa'. This is the D-042 moved-URL case, not a missing fact. PERMITS: the page's verbs are 'live, work or run a business' — RESIDE and WORK recorded. NO STUDY: the page grants no study right anywhere; this route fails the scenario's all-three test on STUDY, exactly like the Virtual Work Residence Visa. WORK_REMOTE NOT RECORDED: 'work ... in the UAE' — the page never states whether a Green Visa holder may work remotely for a non-UAE employer; the 'Freelancers and self-employed' eligible category is suggestive but the page does not state the permitted work's location or employer, and I will not infer it. maxStayDays=1825: 'a renewable residence visa valid for 5 years'; for a residence permit the permitted stay runs the validity. 'Renewable' is not stated as of right, so renewal is not counted. ELIGIBILITY IS CATEGORY-GATED: skilled workers, freelancers/self-employed, investors and business partners — per-person conditions (the page does not state the underlying salary/qualification thresholds; those sit in ICP/MoHRE rules I could not fetch because icp.gov.ae answered HTTP 500 sitewide on the retrieval date). PASSPORT SET '*': no nationality condition appears on the page ('foreign talents'). BASIS IS THE WEAKER ONE: absence of a condition, not an affirmative neutrality statement. effectiveAt is the page's own 'Updated On: 26 Feb 2026' date; the underlying decree (Cabinet Resolution 65/2022) was unreadable — uaelegislation.gov.ae returned HTTP 403 (Cloudflare) to this fetcher. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK — HTTP 200 at the cited (moved) slug .../residence-visa-for-doing-business-in-the-uae, independently fetched 2026-07-18. The page's breadcrumb reads 'Types of visas > Green visa' and its H1 is 'Green Visa', confirming the moved-URL account: the live page at this slug IS the Green Visa page. 'Updated On: 26 Feb 2026', 'Source: ICP'. | seed | seed | 30bfd792-a6bd-4cc9-8abc-55a904415062 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 35 | 68d63c2e-df93-4bcb-959e-f1c7f6237f99 | ae | visa-access | {"route":"Student residence permit (residency without work permission, sponsored by an accredited educational institution)","permits":["RESIDE","STUDY"],"maxStayDays":null,"passportSet":"*"} | NULL | https://www.gdrfad.gov.ae/en/services/f52024e0-b812-11ed-5210-4cd98f768936 | General Directorate of Residency and Foreigners Affairs — Dubai (GDRFA) | 2025-08-13 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE, verbatim from the page on the retrieval date: “A foreign student enrolled in an accredited university, college, educational institution, or research institute in the country is given a residence permit without the requirement that they work. Instead, the educational institution serves as the student's host and guarantor.” The page's own breadcrumb places the service under: “Home > Services > Residency Services > Issuance of Residency without work permission > Issuing a residency permit to a student.” PERMITS: RESIDE and STUDY — the permit is residence for study, sponsored/guaranteed by the institution. NO WORK RIGHT, AND THIS TIME AFFIRMATIVELY: unlike the usual silence-is-not-permission inference, GDRFA's own service taxonomy files this permit under 'Issuance of Residency without work permission' — the issuing authority's category name states the permit carries no work permission. What remains UNANSWERED on any page I could read: whether working remotely for a non-UAE employer requires 'work permission' at all — no fetched official page states it either way, so whether a student could also WORK_REMOTE is unknown, not no. maxStayDays=null, DELIBERATELY: no page I fetched states the permit's duration. The u.ae student page (https://u.ae/en/information-and-services/visa-and-emirates-id/residence-visas/residence-visa-for-studying-in-the-uae, fetched the same day) says only that the applicant must submit 'a certificate issued by the university or institute specifying the duration of the study'; the GDRFA page's Terms/Fees accordions did not render their contents to the fetcher. Search snippets say '1 year, renewable annually' but a snippet is not a source and I did not record it. A null maxStayDays cannot satisfy the ≥180-day requirement, which is the safe direction. SPONSORSHIP CONDITION THE GATE READER MUST SEE: the route requires enrolment at an ACCREDITED university/college/educational institution/research institute (or, per the u.ae page, sponsorship by a resident parent) — the prior fixture already noted Network School would not obviously qualify as the accredited sponsor, and nothing I read changes that. PASSPORT SET '*': 'A foreign student' — no nationality condition on either official page. BASIS IS THE WEAKER ONE: absence of a condition, not an affirmative neutrality statement. effectiveAt is GDRFA's own 'Service details last updated on: 13/08/2025'; the underlying statutory instrument was unreadable (uaelegislation.gov.ae HTTP 403 to this fetcher). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK — HTTP 200 at the cited GDRFA URL, fetched twice independently on 2026-07-18 (DOMContentLoaded and again with waitForNetworkIdle): identical text both times; the Service Details accordions (Requirements / Fees / Terms And Conditions / Additional Information) render only their headings, never their contents, to this fetcher. Breadcrumb verbatim: 'Home > Services > Residency Services > Issuance of Residency without work permission > Issuing a residency permit to a student'. 'Service details last updated on: 13/08/2025'. | seed | seed | 7de3b42a-a233-4f45-bcfe-ba8ccd9390e1 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 36 | 09dd8ec0-fc27-4059-bb0f-8136a0698687 | id | visa-access | UNRESOLVED | NULL | https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G | Direktorat Jenderal Imigrasi, Kementerian Imigrasi dan Pemasyarakatan RI | 2025-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed {"US": [{"route": "E33G Visa Rumah Kedua Pekerja Jarak Jauh (Remote Worker KITAS)", "permits": ["RESIDE", "WORK_REMOTE"], "maxStayDays": 365}, {"route": "E33 Vi and an independent agent re-fetched the source and rejected it. VERIFIER: Page loads (HTTP 200) and the quote is literally present: 'Dengan visa ini, anda dapat / Tinggal di Indonesia untuk melaksanakan tugas dari perusahaan luar negeri. Selama Izin Masuk Kembali masih berlaku, Anda diizinkan untuk masuk dan keluar wilayah Indonesia. Selain itu, Anda juga dapat melakukan aktivitas yang berhubungan dengan wisata, mengunjungi teman atau keluarga'. The page supports ONLY the E33G row: 'Anda dapat tinggal di Indonesia selama 1 tahun' establishes maxStayDays 365, and performing duties for a foreign company establishes RESIDE + WORK_REMOTE. But the claimed value asserts three routes per nationality. The cited URL is the E33G page and is entirely silent on 'E33 Visa Rumah Kedua (Second Home)' (claimed 1825 days, RESIDE/STUDY/WORK_REMOTE) and 'E30B Visa Pendidikan' (claimed 1460 days, RESIDE/STUDY). Two of three routes, replicated across all four nationality blocks (US/EU/IN/CN), have no support from this source. The quote also does not establish that the matrix is identical for CN/IN nationals. Needs per-route citations. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: DIRECT ANSWER TO THE TASKED QUESTION: the E33G does NOT permit STUDY. Its full permitted-activity statement on the immigration authority's own page is quoted verbatim in `quote` and covers exactly three things — residing to perform duties for a FOREIGN company, entering/exiting on a valid Re-entry Permit, and tourism/visiting friends or family. Education is absent. Per the silence rule, absence is not permission, so E33G = RESIDE + WORK_REMOTE only. It is a NEAR-MISS: permits RESIDE/WORK_REMOTE but not STUDY. I corroborated the negative rather than resting on silence from one page. (a) The dedicated E33G page (this sourceUrl) contains ZERO occurrences of 'pendidikan' — I grepped the raw HTML, not a model summary. (b) On the 89-visa omnibus classification page (https://www.imigrasi.go.id/wna/permohonan-visa-republik-indonesia/klasifikasi-visa-tinggal-terbatas-indeks-e33g?golden_visa=0) 82 entries carry an explicit education clause — 'Anda diizinkan mengikuti pendidikan melalui mekanisme Rangkap Kegiatan Izin Tinggal' — attached to investor/golden-visa entries such as E28A. E33G carries no such clause. Indonesia's drafting states education explicitly when it is granted (E30B: 'Menempuh pendidikan di Indonesia pada jenjang sarjana, pascasarjana hingga doktoral'), so its omission from E33G is meaningful, not an oversight. IMPORTANT PROCESS NOTE / TRAP AVOIDED: an initial WebFetch of the omnibus page returned the clause 'melakukan kegiatan yang berhubungan dengan pekerjaan atau mengikuti pendidikan, sepanjang telah dilaporkan ke Imigrasi' as if it described E33G. It does not — that page describes all ~89 visa indexes, and the summarising model misattributed another visa's clause to E33G. Had that been accepted, this gate would have PASSED on a fabricated basis. Raw-HTML verification is why it did not. THE PASS ROUTE IS E33, NOT E33G. E33 Visa Rumah Kedua (Second Home), https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33, states verbatim: 'Melakukan kegiatan yang berhubungan dengan wisata, pembelian barang, serta mengunjungi keluarga dan teman. Selain itu, anda juga dapat melakukan kegiatan bekerja, menempuh pendidikan atau kegiatan lainnya dengan melakukan pelaporan rangkap kegiatan ke kantor imigrasi yang menerbitkan izin tinggal.' — i.e. work AND education on ONE route. Stay: 'Anda diberikan izin tinggal pertama dengan masa berlaku paling lama 5 tahun' → maxStayDays 1825 on one grant (extension to a 10-year total is conditional, so not counted). CAVEATS THE ENGINE SHOULD SURFACE: (1) E33 demands a US$130,000 commitment — 'pernyataan komitmen akan menyimpan dana pada rekening atas nama sendiri pada bank milik negara senilai paling sedikit US$130.000' — a hard economic gate, not a legal bar. (2) 'bekerja' is stated generally; I did NOT resolve whether it subsumes remote work for a foreign employer without a separate work permit (RPTKA/IMTA). Remote work for a foreign employer is the least-restricted reading, but this is the weakest link in the E33 PASS and deserves qualified legal review before relying on it. (3) 'pelaporan rangkap kegiatan' is a reporting formality to the issuing office; I read it as a condition rather than a discretionary grant, but did not source a regulation confirming it is as-of-right. E33G maxStayDays 365 from 'Anda dapat tinggal di Indonesia selama 1 tahun'. E30B maxStayDays 1460 from 'Anda dapat memilih untuk tinggal di Indonesia selama 1 tahun 2 tahun, atau 4 tahun'; E30B is silent on work → no WORK_REMOTE. COHORT CAVEAT — READ BEFORE TRUSTING THE PER-COHORT SPLIT: I did not find per-nationality route data. None of the E33G/E33/E30B pages state any nationality restriction, and E33G explicitly requires no sponsor ('Anda tidak membutuhkan penjamin/sponsor'), so I applied the identical nationality-neutral matrix to all four cohorts. This is an inference from the ABSENCE of a restriction, not a sourced per-cohort finding. Indonesia does operate a 'calling visa' regime for certain listed nationalities (visible as 'NEGARA CALLING VISA' on index C11A); I did not retrieve that list and cannot confirm whether any of US/EU/IN/CN appear on it. If the engine needs cohort-specific confidence, treat that as an open item (listed in `blocked`). effectiveAt 2025-01-01: the current classification rests on 'Keputusan Menteri Imigrasi dan Pemasyarakatan RI Nomor M.IP-08.GR.01.01 tahun 2025 tentang Klasifikasi Visa'; the decree is dated 2025 but its exact promulgation date is not stated on the page, so I used the conservative start of that year. QUOTE THE RESEARCHER OFFERED: “Dengan visa ini, anda dapat Tinggal di Indonesia untuk melaksanakan tugas dari perusahaan luar negeri. Selama Izin Masuk Kembali masih berlaku, Anda diizinkan untuk masuk dan keluar wilayah Indonesia. Selain itu, Anda juga dapat melakukan aktivitas yang berhubungan dengan wisata, mengunjungi teman atau keluarga” | seed | seed | 2a329fd7-e180-49fc-b319-a43a656bffd7 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 37 | 05984129-0381-44da-b76c-02dab477b05a | id | visa-access | {"route":"E33G Visa Rumah Kedua Pekerja Jarak Jauh (Remote Worker)","permits":["RESIDE","WORK_REMOTE","VISIT"],"maxStayDays":365,"passportSet":"*"} | NULL | https://web.archive.org/web/20260520064653/https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G | Direktorat Jenderal Imigrasi, Kementerian Imigrasi dan Pemasyarakatan RI (page content); rendering read via Internet Archive Wayback Machine snapshot of 2026-05-20 | 2025-01-01 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE (verbatim from the archived official page): 'Dengan visa ini, anda dapat / Tinggal di Indonesia untuk melaksanakan tugas dari perusahaan luar negeri. / Selama Izin Masuk Kembali masih berlaku, Anda diizinkan untuk masuk dan keluar wilayah Indonesia. Selain itu, Anda juga dapat melakukan aktivitas yang berhubungan dengan wisata, mengunjungi teman atau keluarga' and 'Anda dapat tinggal di Indonesia selama 1 tahun.' WHAT THE ROUTE PERMITS: residing in Indonesia to perform duties for a company established OUTSIDE Indonesia (RESIDE + WORK_REMOTE), plus tourism and visiting friends/family (VISIT). WHAT IT DOES NOT PERMIT: STUDY — education appears nowhere on the page, and Indonesia's drafting states education explicitly where granted (E30B; and 82 golden-visa index descriptions carry 'Anda diizinkan mengikuti pendidikan melalui mekanisme Rangkap Kegiatan Izin Tinggal' — E33G's own description on the omnibus classification page, archived snapshot 2026-03-14, https://web.archive.org/web/20260314095319/https://www.imigrasi.go.id/wna/permohonan-visa-republik-indonesia/klasifikasi-visa-tinggal-terbatas-indeks-e33g?golden_visa=0 , carries no education clause). NEAR-MISS for the default scenario: permits RESIDE/WORK_REMOTE but not STUDY. Local sales prohibited: 'Anda dilarang ... melakukan penjualan barang atau jasa, terkecuali diperlukan dalam pekerjaannya'. maxStayDays 365 is the single-route per-grant stay ('selama 1 tahun'). The page says the stay permit 'dapat diperpanjang' (may be extended) but publishes no as-of-right extension term, so no extension is counted (per the Vietnam-pass correction: discretionary or unquantified extensions count for nothing). PASSPORTSET '*' BASIS (weaker, absence basis — stated explicitly per PASSPORT.md §1): the page states financial/contractual conditions only — 'Rekening bank yang membuktikan penghasilan berupa gaji atau penghasilan senilai paling sedikit US$60.000 per tahun; Perjanjian kerja dengan perusahaan yang didirikan di luar Wilayah Indonesia' — and 'Anda tidak membutuhkan penjamin/sponsor untuk mengajukan visa ini.' It states NO nationality condition, but I found no affirmative nationality-neutrality statement. Countervailing check performed: the authority's only published per-nationality restriction list (calling-visa subjects: Afganistan, Israel, Korea Utara, Liberia, Nigeria, Somalia — see the passport-exclusions claims in this file) names none of US, any EU/EEA/CH member, IN, or CN. TIER GAP DISCLOSURE (D-043): the live page https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G is OFFICIAL_PRIMARY substance but the host CloudFront-403s this fetcher (fetch fact, 2026-07-18); the URL actually read is the Internet Archive's byte-preserved snapshot, tiered ACADEMIC_OR_INDEX as a non-governmental archival rendering. A checker with browser access should re-fetch the live URL and may then re-tier OFFICIAL_PRIMARY. effectiveAt 2025-01-01: the page's dasar hukum rests on Keputusan Menteri Imigrasi dan Pemasyarakatan RI Nomor M.IP-08.GR.01.01 Tahun 2025 tentang Klasifikasi Visa (plus Permenkumham 22/2023 jo. 11/2024); the decree's promulgation date is not stated on the page, so the conservative start of its year is used, same convention as the 2026-07-17 pass. Not a sourced date. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Cited Wayback snapshot (20260520064653) re-fetched independently, HTTP 200, quotes verified verbatim. Live https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G attempted 2026-07-18: CloudFront 403 to this fetcher — the researcher's reported wall is a shared fetch fact; ACADEMIC_OR_INDEX tiering of the mirror stands. Verification detail: RESIDE+WORK_REMOTE is expressly stated ('melaksanakan tugas dari perusahaan luar negeri' — this route, unlike E33, names remote work for a foreign company in terms); VISIT stated; STUDY correctly absent (page grants none, and the klasifikasi index snapshot cited in the note was also re-fetched — education is granted elsewhere via explicit 'Rangkap Kegiatan Izin Tinggal' clauses, none on E33G). maxStayDays 365 is the per-grant stay ('selama 1 tahun'), correctly distinguished from the 90-day visa validity ('Masa berlaku visa adalah 90 hari') and correctly excluding the unquantified 'dapat diperpanjang'. '*' absence basis honest: page states income (US$60,000/yr), foreign employment agreement, no sponsor — no nationality condition; claim discloses absence-not-affirmation. | seed | seed | 6a292731-1d57-4fb4-885a-6345d480a467 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 38 | 24a8117f-59f3-406d-b6b4-01f2eb3d57f7 | id | visa-access | {"route":"E33 Visa Rumah Kedua (Second Home)","permits":["RESIDE","STUDY","WORK_REMOTE","VISIT"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://web.archive.org/web/20260521070247/https://imigrasi.go.id/wna/daftar-visa-indonesia/E33 | Direktorat Jenderal Imigrasi, Kementerian Imigrasi dan Pemasyarakatan RI (page content); rendering read via Internet Archive Wayback Machine snapshot of 2026-05-21 | 2025-01-01 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | E33 SECOND-HOME FOLLOW-UP (2026-07-18). Task: resolve Q1 ('bekerja' scope / RPTKA) and Q2 (rangkap kegiatan = notification or approval) from PRIMARY regulatory text. VERDICT: the route REMAINS HELD (NEEDS_EVIDENCE). Q1 is answered against the permissive reading at the definitional level; Q2 could NOT be answered from the OPERATIVE (2023/2025) primary text — every host serving the current defining article was walled, timed out, or served the document with its status-change chapter truncated. permits[] and maxStayDays are LEFT UNCHANGED (matching the current operative imigrasi E33 page, which does grant work+study via rangkap kegiatan) precisely BECAUSE the claim stays HELD: I will not correct them down using a superseded instrument, and I cannot raise the claim to pass-eligible while Q2 is open. This claim is not, and its own text says it is not, a resolution that clears the gate. === Q1 — DOES E33 'BEKERJA' COVER REMOTE WORK FOR A FOREIGN EMPLOYER WITHOUT RPTKA/IMTA? === PRIMARY FINDING A (leans AGAINST an as-of-right native work grant). SE Direktur Jenderal Imigrasi Nomor IMI-0820.GR.01.01 Tahun 2022 tentang Visa dan Izin Tinggal Rumah Kedua, BAB I (PENGERTIAN), angka 4, verbatim: 'Visa Tinggal Terbatas Rumah Kedua yang selanjutnya disebut dengan Visa Rumah Kedua adalah Visa TIDAK DALAM RANGKA BEKERJA yang diberikan kepada Orang Asing dan/atau keluarganya yang tinggal menetap di Indonesia selama 5 (lima) tahun atau 10 (sepuluh) tahun setelah memenuhi syarat tertentu.' angka 5, verbatim: 'Izin Tinggal Terbatas Rumah Kedua ... adalah Izin Tinggal terbatas TIDAK DALAM RANGKA BEKERJA ...'. The Second Home visa/permit is DEFINITIONALLY a non-work permit. CAVEAT ON WEIGHT: this SE rests on the SUPERSEDED Permenkumham 29/2021 (its dasar hukum), replaced by Permenkumham 22/2023 jo. 11/2024 and Kepmen M.IP-08.GR.01.01/2025 — so it is FOUNDATIONAL/HISTORICAL context, not the current operative rule. It is quoted because it establishes the programme's non-work character, which the current E33 page relaxes only via the (unresolved, Q2) rangkap-kegiatan add-on. Read verbatim this pass at https://batam.imigrasi.go.id/download_files/2450-1671770819/download/Surat_Edaran_Ditjenim_Nomor_IMI-0820_GR_01_01_Tahun_2022_Tentang_Visa_Dan_Izin_Tinggal_Rumah_Keduapdf (13-page official Ditjenim circular, Plt. Dirjen Imigrasi Widodo Ekatjahjana, ditetapkan 20 Des 2022), via reader proxy. PRIMARY FINDING B (context on the sibling remote-work index, supports that foreign-employer remote work is sponsor-free where it IS granted). Permenkumham 22/2023 (Berita Negara RI No.651/2023), Pasal 33 ayat (2) huruf j, verbatim: 'j. rumah kedua, yang terdiri atas: 1. rumah kedua; 2. keahlian khusus; 3. tokoh dunia; 4. lanjut usia berusia 60 (enam puluh) tahun atau lebih; dan 5. pekerja jarak jauh (remote worker) yang terikat hubungan kerja dengan perusahaan di luar Wilayah Indonesia.' The remote-worker sub-category (E33G) is defined by employment with a company OUTSIDE Indonesia, and Pasal 56-57 grant these categories 'tanpa Penjamin' (without sponsor) for up to 5/10 years. In the ~57 pages of body text obtainable, the regulation contains NO occurrence of 'RPTKA', 'izin kerja', or 'tenaga kerja asing'. STRUCTURAL INFERENCE (mine, disclosed): RPTKA/IMTA is the foreign-worker permit tied to an INDONESIAN employer/sponsor; the Second Home framework is expressly sponsor-free, so it cannot host RPTKA-based local employment — which is why work for a foreign employer is carved out as its own sponsor-free index (E33G) rather than folded into the base Second Home permit. Read verbatim this pass at https://r.jina.ai/https://jdih.sulbarprov.go.id/peraturan/PERMENKUMHAM-22-2023.pdf and https://r.jina.ai/https://kuikk.trisakti.ac.id/wp-content/uploads/sites/128/2024/02/permenkumham-no-22-tahun-2023-1.pdf (two independent gazette copies). Q1 SYNTHESIS: The permitted 'bekerja' on the E33 SECOND-HOME index is NOT a native permission of the permit — the Second Home visa is definitionally 'tidak dalam rangka bekerja', and the current E33 page grants work only as a rangkap-kegiatan add-on. Where the regime DOES grant remote work for a foreign employer, it does so sponsor-free and RPTKA-free (the E33G index). So mapping E33 'bekerja' to WORK_REMOTE-for-a-foreign-employer is the least-restricted and structurally-coherent reading, but it is NOT a native as-of-right feature of the Second Home permit; it depends entirely on the rangkap-kegiatan mechanism (Q2). CORROBORATING (found last pass): the eVisa FAQ for the second-home FAMILY variant states 'Prohibited from working in employment relationship' while allowing 'Study in Indonesia' — i.e. the regime treats Second Home employment-relationship work as prohibited absent a separate authorisation. === Q2 — IS 'PELAPORAN RANGKAP KEGIATAN' A NOTIFICATION (AS-OF-RIGHT) OR A DISCRETIONARY APPROVAL? === UNRESOLVED FROM OPERATIVE PRIMARY TEXT. (1) SE IMI-0820/2022 has NO rangkap-kegiatan mechanism at all — it is a pure non-work visa, so the mechanism is a later (2023+/2025) construct. (2) The obtainable ~57 pages of Permenkumham 22/2023 (front/application chapters, up to BAB II) contain NO occurrence of 'rangkap'; the status-change chapter that would carry 'rangkap jenis kegiatan' is in the document's second half, which every reachable full-text copy served as un-extractable/capped output (both gazette PDFs truncated at the identical ~104.8 KB, i.e. a reader output cap before the chapter, not the document ending). I therefore CANNOT assert 'rangkap is absent from 22/2023' — I did not read the whole instrument. (3) The named mechanism 'Rangkap Kegiatan Izin Tinggal' rests on Kepmen M.IP-08.GR.01.01/2025 (Klasifikasi Visa) jo. Permenkumham 22/2023 as amended by 11/2024; NONE of these operative texts could be read (fetch-fact log in followup-e33-notes.md — BPK Cloudflare-CAPTCHA even via reader; peraturan.go.id / bphn.jdihn.go.id / jdih.imigrasi.go.id connection-closed; imigrasi.go.id CloudFront-403; regional Kanim WAF/maintenance; regulasip 404; hukumonline paywall). LEANING SIGNALS (secondary, NOT sufficient to resolve): the authority's own klasifikasi page drafts the mechanism perfectively — 'sepanjang telah melalui mekanisme Rangkap Kegiatan Izin Tinggal' ('provided one HAS GONE THROUGH the mechanism') — and the analogous ITAS-Pekerja 'rangkap jabatan' path is documented as a photo+interview process with pickup 7 working days after; both point toward an APPROVAL/issuance step rather than a courtesy notification. But these describe adjacent constructs, not the operative Second-Home rangkap-kegiatan article, so Q2 stays open. === NET FOR THE GATE === Does E33 permit RESIDE+STUDY+WORK_REMOTE AS-OF-RIGHT? On the reachable evidence: NO, not established. RESIDE is inherent. STUDY and WORK_REMOTE are add-ons available only 'dengan melakukan pelaporan rangkap kegiatan'; the base permit is foundationally 'tidak dalam rangka bekerja'; and whether the add-on is as-of-right (notification) or discretionary (approval) is UNRESOLVED. The balance of newly-obtained primary/official evidence leans AGAINST a clean as-of-right PASS, not toward one. Route stays HELD. UPGRADE PATH unchanged: obtain, via browser or human, the operative Permenkumham 22/2023 (status-change chapter) as amended by 11/2024, or Kepmen M.IP-08.GR.01.01/2025, and read the article defining 'Rangkap Kegiatan Izin Tinggal'. If it is a notification as-of-right AND scopes 'bekerja' to include foreign-employer remote work, raise to CORROBORATED and Indonesia PASSes; if it is a discretionary approval or scopes work to Indonesian-sponsored employment only, CORRECT permits down (drop WORK_REMOTE and/or STUDY) and Indonesia loses its only PASS candidate — a real result. TIER (D-043): the route grant is the imigrasi E33 page (OFFICIAL_PRIMARY substance) read via Wayback (fixture capture, not re-fetched this pass); the Q1 evidence was read via r.jina.ai reader proxy of a provincial JDIH (jdih.sulbarprov.go.id) and university (trisakti.ac.id) gazette copies and of an official Kanim download (batam.imigrasi.go.id) — substance OFFICIAL_PRIMARY, rendering path non-authoritative, so tiered ACADEMIC_OR_INDEX with this disclosure; a checker with browser access should re-fetch peraturan.bpk.go.id/details/272044 (22/2023) and /Details/285156 (11/2024) live and may re-tier OFFICIAL_PRIMARY. MERGE NOTE, 2026-07-18 (targeted follow-up): the value and NEEDS_EVIDENCE hold are UNCHANGED — the follow-up resolved neither deciding question cleanly, and correctly declined both to correct WORK_REMOTE down off a superseded circular (SE 2022 on PMK 29/2021) and to upgrade while Q2 is open. Net movement is AGAINST a clean PASS, not toward one: the 2022 SE calls Second Home a non-work visa verbatim, and Permenkumham 22/2023 Pasal 33(2)(j) shows the sponsor-free foreign-employer remote-work permission lives in the SIBLING index E33G, so 'bekerja' is an add-on to Second Home rather than its native grant. No verifier pass was run: the follow-up changed no gate value and added no pass-eligible claim, so there is nothing for an adversarial check to refute — E33 stays exactly where the first verifier left it, below the gate. UPGRADE PATH unchanged: browser/human retrieval of Permenkumham 22/2023 jo. 11/2024 and Kepmen M.IP-08.GR.01.01/2025 to resolve Q2 (notification vs approval); a for-the-route answer flips Indonesia to the product's first visa PASS. | seed | seed | 6e507dc1-b2fc-47b1-8af2-a4a2a01d7f43 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 39 | 0b0dee8f-a311-445f-8b0b-9db1e63e3b19 | id | visa-access | {"route":"E30B Visa Pendidikan Tinggi (Higher Education KITAS)","permits":["RESIDE","STUDY","VISIT"],"maxStayDays":1460,"passportSet":"*"} | NULL | https://web.archive.org/web/20260520071133/https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E30B | Direktorat Jenderal Imigrasi, Kementerian Imigrasi dan Pemasyarakatan RI (page content); rendering read via Internet Archive Wayback Machine snapshot of 2026-05-20 | 2025-01-01 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE (verbatim from the archived official page): 'Menempuh pendidikan di Indonesia pada jenjang sarjana, pascasarjana hingga doktoral. Anda juga dapat membawa anggota keluarga untuk tinggal di wilayah Indonesia ... Selain itu, Anda juga dapat melakukan aktivitas yang berhubungan dengan wisata, mengunjungi teman atau keluarga' and 'Anda dapat memilih untuk tinggal di Indonesia selama 1 tahun 2 tahun, atau 4 tahun dihitung sejak tanggal kedatangan.' WHAT THE ROUTE PERMITS: residing to pursue degree-level education (bachelor through doctoral) — RESIDE + STUDY — plus tourism/visits (VISIT). WHAT IT DOES NOT PERMIT: work of any kind — the page grants none, and its Ketentuan lain prohibits 'melakukan pekerjaan yang tidak sesuai dengan izin tinggal, melakukan penjualan barang atau jasa'. Per the silence rule, no WORK_REMOTE. NEAR-MISS for the default scenario: permits RESIDE/STUDY but not WORK_REMOTE. maxStayDays 1460 = the longest selectable per-grant stay (4 years, chosen at application; 1/2/4-year options are all stated as-of-right on the page). Extensions exist ('dapat diperpanjang') with no published as-of-right term — not counted. SPONSOR CONDITION (not a nationality condition): 'Anda membutuhkan penjamin/sponsor untuk mengajukan visa ini' — 'bukti penjaminan dari Penjamin yang merupakan perorangan atau institusi pendidikan di mana WNA menempuh pendidikan' plus 'Surat penerimaan dari institusi pendidikan'. This is an admission/sponsorship gate, meaning the route qualifies a member only once enrolled with an Indonesian institution — decision-relevant for a school-shaped project and disclosed here rather than hidden. PASSPORTSET '*' BASIS (weaker, absence basis — stated explicitly): no nationality condition appears on the page; no affirmative neutrality statement found. Countervailing check: the published calling-visa list (6 states) contains none of US / EU/EEA/CH members / IN / CN. TIER GAP DISCLOSURE (D-043): live host CloudFront-403s this fetcher; URL read is the Internet Archive snapshot, tiered ACADEMIC_OR_INDEX; substance is the issuing authority's own page. effectiveAt 2025-01-01 per the Kepmen M.IP-08.GR.01.01/2025 start-of-year convention, disclosed as unsourced to the day. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Cited Wayback snapshot (20260520071133) re-fetched independently via the id_ raw variant of the SAME capture timestamp after the toolbar-rendered URL timed out; HTTP 200, quotes verified verbatim. Live host CloudFront-403s this fetcher (confirmed on sibling URLs same session). Verification detail: RESIDE+STUDY+VISIT match; no work of any kind granted and 'Anda dilarang ... melakukan pekerjaan yang tidak sesuai dengan izin tinggal, melakukan penjualan barang atau jasa' present — WORK_REMOTE correctly withheld under the silence rule. maxStayDays 1460 = the 4-year selectable option, stated as a choice at application; extensions unquantified and correctly not counted. Sponsor/admission gate ('Anda membutuhkan penjamin/sponsor'; 'Surat penerimaan dari institusi pendidikan') verified verbatim and honestly disclosed. No nationality condition on page; '*' absence basis matches. | seed | seed | b30b09d4-9c60-4ca4-9a17-18365b75a0e9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 40 | 77c6e0e9-d172-436c-ac40-a5a9e05bda19 | id | passport-exclusions | {"status":"PRIOR_APPROVAL_REQUIRED","instrument":"Calling Visa regime (Daftar Subjek Calling Visa), Direktorat Jenderal Imigrasi","passportSet":"AF"} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE (verbatim, the page's complete list under 'Daftar Subjek Calling Visa / Daftar Negara, Pemerintah Dari Daerah Administrasi Khusus Suatu Negara, dan Entitas Tertentu'): 'Afganistan, Israel, Korea Utara, Liberia, Nigeria, Somalia'. Nationals of the listed states may only apply through the calling-visa procedure, which requires special approval — a published per-nationality restriction, PRIOR_APPROVAL_REQUIRED, not ENTRY_BARRED. Fetched live and direct (HTTP 200) from a Kanim page of the Directorate General; the DG's own main-site list page (imigrasi.go.id/wna/daftar-negara-voa-bvk-calling-visa/daftar-negara-calling-visa) CloudFront-403s this fetcher and its Wayback snapshot timed out twice — refusals recorded as fetch facts, not absence. The underlying instrument and its date are not stated on the page; effectiveAt is conservatively set to the retrieval date and is NOT a sourced promulgation date. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200, page read to its end (footer/address after Somalia; not truncated). Afganistan is on the list. Caveat recorded: the cited page lists names under the 'Daftar Subjek Calling Visa' heading and does not itself define the calling-visa procedure; the PRIOR_APPROVAL_REQUIRED status rests on the regime's meaning (DG Imigrasi press/FAQ materials describe an inter-ministerial assessment team for calling-visa applications), not on text of this page. List membership and instrument name are on-page; status enum is a faithful mapping, not a quotation. | seed | seed | 4ec10a7e-1543-42a3-8816-f5bcef57c5e0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 41 | 8db6bcb2-34b0-411c-873a-1c825261bae3 | id | passport-exclusions | {"status":"PRIOR_APPROVAL_REQUIRED","instrument":"Calling Visa regime (Daftar Subjek Calling Visa), Direktorat Jenderal Imigrasi","passportSet":"IL"} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Same source and quote as the AF claim: Israel is on the published calling-visa list (special approval required). Context lead NOT claimed here: Indonesia is reported to have no diplomatic relations with Israel (PASSPORT.md §5 diplomatic-relations subject); no primary source for the relations fact was fetched this pass, so no diplomatic-relations claim is emitted — recorded as a lead in researcher-notes.md, not as a value. effectiveAt = retrieval date, disclosed as unsourced. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200. Israel is on the list. Same caveat as claimIndex 3 on the status enum. The claim correctly refrains from asserting a diplomatic-relations fact it did not source. | seed | seed | 5b0dbfd1-68d7-49a4-88ed-49514c2912d0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 42 | a6abf88f-5689-477a-a395-6e5fa87b1c4e | id | passport-exclusions | {"status":"PRIOR_APPROVAL_REQUIRED","instrument":"Calling Visa regime (Daftar Subjek Calling Visa), Direktorat Jenderal Imigrasi","passportSet":"KP"} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Same source and quote as the AF claim: 'Korea Utara' (North Korea) is on the published calling-visa list (special approval required). effectiveAt = retrieval date, disclosed as unsourced. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200. 'Korea Utara' (North Korea) is on the list. Same caveat as claimIndex 3 on the status enum. | seed | seed | 16b2971a-d527-4ace-a1fb-6002664be577 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 43 | f68807e9-1fc5-4158-ba14-1f0698d1c8f8 | id | passport-exclusions | {"status":"PRIOR_APPROVAL_REQUIRED","instrument":"Calling Visa regime (Daftar Subjek Calling Visa), Direktorat Jenderal Imigrasi","passportSet":"LR"} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Same source and quote as the AF claim: Liberia is on the published calling-visa list (special approval required). effectiveAt = retrieval date, disclosed as unsourced. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200. Liberia is on the list. Same caveat as claimIndex 3 on the status enum. | seed | seed | 9edcaa81-cebc-4265-acc0-543e50bd7ba0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 44 | bd06421b-cd2f-4cd9-8a80-180b6e963ceb | id | passport-exclusions | {"status":"PRIOR_APPROVAL_REQUIRED","instrument":"Calling Visa regime (Daftar Subjek Calling Visa), Direktorat Jenderal Imigrasi","passportSet":"NG"} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Same source and quote as the AF claim: Nigeria is on the published calling-visa list (special approval required). effectiveAt = retrieval date, disclosed as unsourced. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200. Nigeria is on the list. Same caveat as claimIndex 3 on the status enum. | seed | seed | c9f86aec-a926-48e4-88df-29af5037d3f3 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 45 | 92c2e1fe-731d-4367-af9b-3d2a43772c6a | uz-tashkent-city | local-additionality | 3 | rubric 0-5 | https://yuz.uz/en/news/xalqaro-taielim-standartlari-asosida-kadrlar-taierlas-biica | State Institution 'Editorial Office of the newspapers Yangi O‘zbekiston and Pravda Vostoka' (yuz.uz), reporting the President's press service | 2026-07-10 | 2026-07-20T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | DOCUMENT_EXTRACTION | NULL | ANCHORS (contracts/rubric.md §2; D-035: 'economic weakness alone is not opportunity'; '3+ requires a published plan or a statistic, not an impression', else cap at 2). Reference scale: the 150-bed / ~300-person node. VERIFIER: CONFIRMED, and it ruled independently on 3 vs 2 (evidence/tashkent-city/verifier-verdicts.md §2) — every quoted sentence was re-fetched and found literally present on yuz.uz (HTTP 200) and the gazeta.uz corroboration (HTTP 200). THE FACTS, each with a resolving URL. (1) A PUBLISHED PLAN NAMES THIS ACTIVITY IN THIS CITY — VERBATIM from yuz.uz, 10 Jul 2026: 'The presentation also reviewed a proposal to establish an International University Town in Tashkent, which will be home to leading domestic and foreign higher education institutions.' 'To implement the project, it is proposed to draw on the experience of Education City in Qatar, Dubai International Academic City in the United Arab Emirates, and Incheon Global Campus in the Republic of Korea.' 'Plans also call for attracting branches of leading foreign universities to the university town, constructing academic campuses and student dormitories, simplifying licensing procedures, providing grants and scholarships, introducing customs and tax incentives, and partially compensating royalty payments.' On status: 'The President approved the proposals and instructed the responsible officials to ensure the phased expansion of integrated educational programs.' Corroborated at https://www.gazeta.uz/en/2026/07/11/international-university-campus/ : 'The International University Campus is planned to be opened in Tashkent, housing leading foreign and domestic universities.' (2) THE GAP IS DOCUMENTED AND QUANTIFIED — VERBATIM from the same page: 'higher education enrollment has increased from 9 to 44 percent, while the number of students has grown from 270,000 to 1.7 million.' VERIFIER CORRECTION APPLIED: the researcher additionally asserted that 'the Uzbekistan-2030 Strategy target is at least 50 percent', i.e. a ~6-point shortfall. That target is NOT on the cited page (the verifier read it in full: it lists a top-500 universities goal, a 100-programme accreditation goal and a 30% certification goal, and no 50% enrollment target). A WebSearch suggests the target is real, but a search summary is not a retrieval — so the claim is STRUCK rather than carried uncited. The anchor does not depend on it. (3) THE MARKET IS NOT SATURATED — same source: '157 university academic programs have received international accreditation', and British integrated programmes at Inha University Tashkent and Tashkent University of Architecture and Civil Engineering with only '371 students currently studying' under them. ANCHOR MAPPING -> 3: anchor 3 is 'District-material: would rank among the larger tenants/employers in its district, OR fills a gap a published local plan names.' The SECOND limb is the one claimed (named explicitly, so it does not degrade on the two-part-claim rule): a published, presidentially-approved plan names an international education cluster IN TASHKENT and is recruiting this activity class with licensing, tax and customs incentives. THE DIRECT CONTRAST WITH DUBAI, which capped at 2: Dubai's D33 named higher education but DIAC (27 universities, 27,500 students) already SATURATED the provision, so a node filled no unmet gap (D-035, marginal contribution not wealth). Tashkent is the inverse — the state is offering incentives to attract providers. WHY NOT 4/5: anchor 4 needs a top-10-in-category city-wide ranking with documented local demand; anchor 5 needs the strategy to seek THIS activity for a region-largest effect. Neither holds — no sector statistic ranks a ~300-person node top-10 in a city of 3,164,030 (Tashkent infobox, 1 October 2025, citing the national statistics committee), no demand study exists, and the plan is at PROPOSAL-APPROVED stage: no gazetted instrument, no decree number, no allocated site. Presidential approval of a proposal is not a statute and is not treated as one. THE HONEST WEAKNESS, stated rather than hidden, and the verifier sharpened it: the gap the plan names is DOMESTIC youth enrollment, whereas this node houses ~300 largely international members; and the plan names BRANCHES OF FOREIGN UNIVERSITIES, which a residential co-living/education node is not either. So both limbs of 'fit' are looser than a 3 suggests. What holds 3 up is CONSISTENCY WITH THE BALI PRECEDENT, which this repo verified at 3 on an equally loose activity match ('education and technology park'). Ruling Tashkent to 2 on strictness while Bali sits at 3 on the same looseness would be incoherent — so this is a CONTRACT question, not a Tashkent question: if the owner tightens what 'fills a gap a published local plan names' requires by way of activity-class fit, BALI AND TASHKENT MOVE TOGETHER. Recorded, not scored: the instrument here is a state press report of an approved proposal, materially weaker than Bali's gazetted PP 23/2023 which earned the same 3. SCOPE DISCIPLINE: the 'New Tashkent' education cluster (New Uzbekistan University and Tashkent State Pedagogical University campuses, https://president.uz/en/lists/view/7135) is being built in Tashkent REGION, not the city, so it is NOT counted toward this CITY score — the verifier independently confirmed that exclusion. | seed | seed | 39fa889d-29ff-4a8f-8928-ec9875cd8c20 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 46 | d5b852df-7761-4c28-ae81-c1a58a17f5dd | id | passport-exclusions | {"status":"PRIOR_APPROVAL_REQUIRED","instrument":"Calling Visa regime (Daftar Subjek Calling Visa), Direktorat Jenderal Imigrasi","passportSet":"SO"} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Same source and quote as the AF claim: Somalia is on the published calling-visa list (special approval required). effectiveAt = retrieval date, disclosed as unsourced. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200. Somalia is on the list (the final entry; the page ends after it). Same caveat as claimIndex 3 on the status enum. | seed | seed | 01928563-9a02-4adb-bfb9-6b51a565ad24 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 47 | 08275f14-18b9-4e19-b69b-4aa0f00d423a | id | passport-exclusions | {"passportSet":"*","exclusionsEnumerated":true} | NULL | https://jakartapusat.imigrasi.go.id/layanan/warga-negara-asing-wna/daftar-subjek-voa-bvk-calling-visa/daftar-subjek-calling-visa | Kantor Imigrasi Kelas I Non TPI Jakarta Pusat, Direktorat Jenderal Imigrasi | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | ENUMERATION, WITH ITS EXACT SCOPE STATED — DO NOT OVER-READ. The cited page presents the authority's complete published calling-visa subject list, and I read it to the end: exactly 6 entries (Afganistan, Israel, Korea Utara, Liberia, Nigeria, Somalia), each emitted as its own PRIOR_APPROVAL_REQUIRED claim in this file. SCOPE LIMITS: (1) this enumerates the CALLING-VISA (special-approval) category only; I found no published ENTRY_BARRED nationality list for Indonesia, and per AGENTS.md §4 I do not assert that none exists — nobody-looked is not the same as none. (2) listSize (PASSPORT.md §5's denominator, e.g. 'of 199 nationalities') is deliberately OMITTED: the page states no denominator of nationalities considered, so no 'N of M' rendering and no percentage is licensed by this claim. (3) None of the four scenario passport sets (US, EU/EEA/CH members, IN, CN) appears on the list — that is the countervailing check backing the '*' basis of the three route claims. effectiveAt = retrieval date, disclosed as unsourced (the underlying instrument's date is not on the page). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live fetch 2026-07-18, HTTP 200, textLength 1244, truncated:false — I independently read the page to its end: exactly six entries, no more. The publisher is a Kanim (regional office) of the Directorate General republishing the DG list — OFFICIAL_SECONDARY tiering is correct. Scope limits verified honest: enumerates the calling-visa category only; asserts no ENTRY_BARRED list and no denominator (listSize deliberately omitted, so no percentage is licensed); no routesEnumerated claim exists in the file and none of the ten claims smuggles a bar. Context from the open web (not from the cited page): the 2020-era calling-visa list had 8 states (including Guinea and Cameroon) — the list demonstrably changes over time, which supports the claim's disclosure that effectiveAt is the retrieval date, not a promulgation date. | seed | seed | e8f70e10-30d7-478d-bc4c-550a955e076c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 48 | 290f7052-bd9d-4ee4-9e03-6746dcd40b70 | id | foreign-title-statute | true | NULL | https://peraturan.bpk.go.id/Download/154522/PP%20Nomor%2018%20Tahun%202021.pdf | Pemerintah Republik Indonesia (via JDIH Badan Pemeriksa Keuangan RI) | 2021-02-02 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | TRUE as a matter of national statute. PP 18/2021 Pasal 49(2) lists the subjects who may hold 'hak pakai dengan jangka waktu' (right of use, for a term) and letter (c) expressly includes 'badan hukum asing yang mempunyai perwakilan di Indonesia' — a foreign legal entity holding a representation in Indonesia. Pasal 51(1) confirms such land may be Tanah Negara, Tanah hak milik, or Tanah Hak Pengelolaan. SCOPE AND LIMITS (do not over-read this TRUE): (1) It is Hak Pakai, not freehold. Hak Milik (freehold) remains barred to foreign entities. (2) Hak Guna Bangunan (HGB) is NOT open to a foreign entity as such — Pasal 49(2)(b) reserves the Indonesian-law-entity route, so foreign capital typically reaches HGB via a PT PMA (a company incorporated under Indonesian law and domiciled in Indonesia), which is a different legal person from the foreign entity itself. (3) The gate asks about national statute, not this building, and that is what is answered here. (4) The 'perwakilan di Indonesia' condition is a real precondition: a foreign entity with no Indonesian representation does not qualify. SOURCE-QUALITY DISCLOSURE: the BPK-hosted PDF is a scanned image and the text was OCR-extracted with pdfminer. The OCR contains visible artifacts ('darr' for 'dan', 'dimakstrd' for 'dimaksud', 'Februai2O2T' for 'Februari 2021'). The quote is reproduced exactly as extracted, sic errors included, so a re-fetching agent can reconcile; the substantive letter (c) text is clean and unambiguous. effectiveAt 2021-02-02 is sourced from the document itself: 'Diundangkan di Jakarta pada tanggal 2 Februari 2021'. I first tried https://peraturan.go.id/files/pp18-2021bt.pdf, which hung and timed out. SOURCE QUOTE, verbatim from the page on the retrieval date: “Pasal 49 (1) Hak pakai terdiri atas: a. hak pakai dengan jangka waktu; darr b. hak pakai selama dipergunakan. (2) Hak pakai dengan jangka waktu sebagaimana dimakstrd pada ayat (1) huruf a diberikan kepada: a. Warga Negara Indonesia; b. badan hukum yang didirikan menurut hukum Indonesia dan berkedudukan di Indonesia; c. badan hukum asing yang mempunyai perwakilan di Indonesia; d. badan keagamaan dan sosial; dan e. Orang Asing.” | seed | seed | c5ded2fa-2e64-4ef6-bc72-ac83735a495c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 49 | d7bcb8d0-73a4-4849-a3b9-55c86f2c41e6 | id | political-stability | 55.4895207 | WGI percentile rank (0-100, higher better) | https://api.worldbank.org/v2/country/IDN/indicator/GOV_WGI_PV.SC?format=json&source=3&date=2024 | World Bank — Worldwide Governance Indicators | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | UNVERIFIED | OFFICIAL_API | NULL | WGI 'Political Stability - Governance score (0-100)' for Indonesia, reference year 2024 (latest available; WGI source last updated 2026-03-18). Quote is the verbatim JSON record returned by the API. Retrieval note for future passes: the familiar indicator id PV.PER.RNK is now ARCHIVED (it resolves to source 57, 'WDI Database Archives', and returns 'The indicator was not found. It may have been deleted or archived'). The live series is GOV_WGI_PV.SC under source=3, which must be passed explicitly. Trend is flat: 2020 56.67, 2021 54.80, 2022 52.87, 2023 55.41, 2024 55.49. Single-source (the issuing body), so UNVERIFIED rather than CORROBORATED. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator": {"id": "GOV_WGI_PV.SC", "value": "Political Stability - Governance score (0-100)"}, "country": {"id": "ID", "value": "Indonesia"}, "countryiso3code": "IDN", "date": "2024", "value": 55.4895207, "unit": "", "obs_status": "", "decimal": 1}” | seed | seed | 2b73f4e6-072a-41d5-bca0-686acf082839 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 50 | 38da2646-a5e6-4edc-ab16-eb6b77e98c37 | id | rule-of-law | 53.4090118 | WGI percentile rank (0-100, higher better) | https://api.worldbank.org/v2/country/IDN/indicator/GOV_WGI_RL.SC?format=json&source=3&date=2024 | World Bank — Worldwide Governance Indicators | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | UNVERIFIED | OFFICIAL_API | NULL | WGI 'Rule of Law - Governance score (0-100)' for Indonesia, reference year 2024 (latest available). Quote is the verbatim JSON record. Same retrieval caveat as political-stability: RL.PER.RNK is archived; use GOV_WGI_RL.SC with source=3. Trend flat: 2020 54.42, 2021 54.49, 2022 54.74, 2023 55.56, 2024 53.41. NOTE: this is the World Bank WGI rule-of-law measure, deliberately used in place of WJP Rule of Law, which is on the prohibited non-commercial list. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator": {"id": "GOV_WGI_RL.SC", "value": "Rule of Law - Governance score (0-100)"}, "country": {"id": "ID", "value": "Indonesia"}, "countryiso3code": "IDN", "date": "2024", "value": 53.4090118, "unit": "", "obs_status": "", "decimal": 1}” | seed | seed | a6a40fc7-5d37-43b5-8909-73005fa3d7af | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 51 | 43b206f7-9b9c-444d-bc57-01290db3ae2e | id | corruption | 34 | TI CPI score (0-100, higher better) | https://images.transparencycdn.org/images/CPI2025_Results.xlsx | Transparency International | 2025-12-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | CPI 2025: Indonesia score 34, rank 109 of 182. Taken from TI's own official results workbook (linked from https://www.transparency.org/en/cpi/2025/index/idn), sheet 1 'Corruption Perceptions Index 2025: Global scores'. I verified the column semantics against the header row rather than assuming position — header row 4 reads ['Country / Territory', 'ISO3', 'Region', 'CPI 2025 score', 'Rank', 'Standard error 2025', 'Number of sources', 'Lower CI', 'Upper CI'], so 34 is the SCORE and 109 is the RANK. The quote is the Indonesia data row as extracted, in that column order. 90% CI is 30.16-37.84 across 9 sources, i.e. the score is not precise to the unit. The HTML country page itself does not contain the score (it is JS-rendered and the raw HTML carries only a country selector), which is why the workbook is cited directly. Sheet 1 is marked 'EMBARGOED UNTIL 10.02.2026', indicating publication 2026-02-10 for reference year 2025; effectiveAt is set to the reference year end. SOURCE QUOTE, verbatim from the page on the retrieval date: “Indonesia IDN AP 34 109 2.3438409999999998 9 30.156099999999999 37.843899999999998” | seed | seed | 4d01122e-4993-4d77-a0d1-6b0558779235 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 52 | e7c62694-8ee6-4cd7-b9f0-71419eeb9b0c | uz-tashkent-city | disaster-exposure | 3.6 | INFORM Risk 0-10 | https://data360api.worldbank.org/data360/data?DATABASE_ID=DRMKC_INFORM&REF_AREA=UZB | World Bank Data360 (relaying JRC/DRMKC INFORM Risk Index) | 2025-01-01 | 2026-07-20T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | OWNER RULING (D-044): INFORM as a COARSE SCREEN, tier-flagged, and it is NATIONAL. VERIFIER: CONFIRMED — re-fetched (HTTP 200) and all four figures checked at 'LATEST_DATA': true, 'TIME_PERIOD': '2025'. VERBATIM from the World Bank Data360 keyless JSON API: INDICATOR 'INFORM_OVRL', REF_AREA 'UZB', UNIT_MEASURE '0_TO_10' carries "OBS_VALUE": "3.6". The same call returns the 2025 sub-scores this composite launders, all LATEST_DATA true: INFORM_HA (Hazard & Exposure) 4.9, INFORM_VU (Vulnerability) 2.9, INFORM_CC (Lack of Coping Capacity) 3.4. The series is stable over a decade (3.9 in 2016-17 easing to 3.5-3.6 by 2023-25), so 3.6 is not a one-year artefact. COARSE-SCREEN CAVEAT (DATA-SOURCES.md §5): INFORM 'is already a score' and launders its sub-scores; this is a COUNTRY-level composite used as a coarse screen, NOT a reviewable CITY observation — attached to Tashkent as INHERITED-NATIONAL, and it says nothing Tashkent-specific. THE FLATTENING IS MATERIAL HERE: Tashkent's dominant local hazard is SEISMIC and it is demonstrated — VERBATIM from https://en.wikipedia.org/wiki/Tashkent (both agents): 'Much of Tashkent was destroyed in the 1966 Tashkent earthquake' and 'The city is located in an active tectonic area, suffering from numerous tremors and occasional earthquakes.' VERIFIER CORRECTION APPLIED: it is NOT true that INFORM ignores seismic — its Hazard & Exposure component (4.9, notably HIGHER than the composite) does incorporate earthquake hazard at NATIONAL resolution. The correct and still-damning claim is narrower: INFORM does not express TASHKENT-SPECIFIC seismic exposure, and the roll-up to 3.6 dilutes a 4.9 hazard sub-score. This criterion must NOT be read as covering Tashkent's earthquake exposure; a dedicated seismic observation (GEM/USGS PGA) is OWED, and BOTH city verifiers independently concluded seismic hazard has no home in the CITY criterion set at all. TIER NOTE: the URL read is the Data360 mirror (MULTILATERAL) of the primary JRC/DRMKC INFORM index. effectiveAt = the 2025 release year (start-of-year convention; no exact release date on the API record). | seed | seed | 4730cb5a-b17e-4975-a8da-4ca52be007c2 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 53 | efe467a9-c38a-48f3-8cf0-bd52fe95b0cb | uz-tashkent-city | flood-sea-level | 1 | rubric 0-5 | https://www.thinkhazard.org/en/report/39698-uzbekistan-tashkent-city/FL | ThinkHazard! — Global Facility for Disaster Reduction and Recovery (GFDRR) / World Bank | 2026-07-20 | 2026-07-20T00:00:00.000Z | MULTILATERAL | CORROBORATED | DOCUMENT_EXTRACTION | NULL | SCALE DIRECTION (contracts/rubric.md §8, Risk axis, LOWER is better): 0 = inland/elevated & safe; 5 = reclaimed / below 2 m / documented inundation at this site. HIGHER rubric = MORE exposed. VERIFIER: CONFIRMED, and it ruled independently that the criterion IS applicable to a doubly-landlocked city (evidence/tashkent-city/verifier-verdicts.md §3): anchors 3-5 are coastal/reclamation/subsidence constructs that cannot bind here, but anchors 0, 1 and 2 turn on RETURN PERIOD and ELEVATION, which are hydrological and transfer to an inland riverine city without strain. That is the limb scored. THE FACTS, each with a resolving URL. (1) ELEVATION AND SETTING — VERBATIM from https://en.wikipedia.org/wiki/Tashkent infobox: 'Elevation 455 m (1,493 ft)', corroborated independently by OurAirports for TAS ('Field elevation: 1,417 ft / 432 m MSL'). Uzbekistan is doubly landlocked; there is no coastal or sea-level exposure of any kind. (2) THIS IS NOT A DRY-SITE FREEBIE — VERBATIM: 'Tashkent sits at the confluence of the Chirchiq River and several of its tributaries, and is built on deep alluvial deposits up to 15 m (49 ft) in depth.' A confluence site on alluvium is exactly where riverine flood hazard would live if it lived anywhere. (3) THE RESOLVING FLOOD MAP — VERBATIM from ThinkHazard! for the administrative unit 'Tashkent city' (HTTP 200, re-fetched by the verifier): 'River flood — Hazard level: Low' and 'In the area you have selected (Tashkent city) river flood hazard is classified as low based on modeled flood information currently available to this tool. This means that there is a chance of more than 1% that potentially damaging and life-threatening river floods occur in the coming 10 years (return period of c. 1 in 1000 years).' The companion urban-flood page (/UF) returns 'Urban flood — Hazard level: Low'. VERIFIER CORRECTION APPLIED: the researcher wrote that /UF returns the SAME classification including the 1-in-1000 return period, implying an independent urban-surface computation. It does not — /UF's explanatory sentence is verbatim RIVER-flood boilerplate. So /UF establishes a LOW URBAN CLASSIFICATION and nothing more; the return period rests on the fluvial page alone. The anchor does not move (Low on both layers is what the score rests on). ANCHOR MAPPING -> 1: anchor 1 is 'Outside the 1-in-500-year zone, no coastal exposure.' Both conjuncts hold: c. 1-in-1000 is a LONGER (rarer) return period than 1-in-500, and a doubly-landlocked city at 455 m has no coastal exposure. WHY NOT 0 — anchor 0 requires BOTH '>20 m above sea level' AND 'outside any mapped flood zone'. The first is met 20-fold; the second FAILS, and the verifier found the detail that clinches it: ThinkHazard's own legend carries FOUR classes — High, Medium, Low, Very low — and Tashkent is returned as 'Low', not 'Very low', on both layers. The tool had a lower class available and did not assign it, so a positively-mapped, non-minimal hazard cannot be rounded to 'outside any mapped flood zone'. rubric.md §8's 'missing information is not safety' applies in the converse direction. WHY NOT 2: anchor 2 is 'Within a 1-in-500-year zone, or 10-20 m elevation with a watercourse nearby' — the return period is longer than 1-in-500 and 455 m is far outside the 10-20 m band; the 'watercourse nearby' clause is true but only bites in combination with that elevation. CAVEATS BOUNDING WHAT THIS LICENSES: (a) ThinkHazard is a coarse GLOBAL screening tool and says so — VERBATIM: 'The hazard levels and guidance given in ThinkHazard! do not replace the need for detailed natural hazard risk analysis and/or expert advice.' Anchor 1's sourcing column asks for a 'National flood map', so the score rests on a coarser instrument than the anchor nominally names. (b) It warns 'flood hazard can change dramatically over short distances' — this is a CITY-general call and a site on the Chirchiq floodplain or a canal bank needs its own assessment. (c) NO DATA VINTAGE IS STATED for the underlying hazard layer, and the schema requires a date (effective_at is NOT NULL), so effectiveAt = 2026-07-20 = the date the classification was READ. Read it as exactly that and nothing more: it dates OUR retrieval of the tool's current published classification, NOT the age of the model behind it, which ThinkHazard does not disclose. The consequence is that this row's freshness decay measures how recently we re-checked the tool, not how recent its data is — a checker who learns the layer's vintage should replace this date with it. (d) Climate signal, VERBATIM: 'The present hazard level may increase in the future due to the effects of climate change.' (e) LICENCE, checked because it matters commercially and it is CLEAN, unlike the air-quality source: 'Classified hazard levels are licenced under CC-BY'. (f) SCOPE FLAG: this criterion covers FLOOD and SEA LEVEL only. Tashkent's serious natural hazard is SEISMIC (1966), and a low flood score must NOT be read as a low overall natural-hazard score. | seed | seed | 30bc7c3f-2c66-44fa-a345-296515752c40 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 54 | eb7b4c51-863e-41a5-b814-410490d2cee9 | id | currency-inflation | 1.9 | CPI %/yr | https://www.imf.org/external/datamapper/api/v1/PCPIPCH/IDN | International Monetary Fund — World Economic Outlook (DataMapper) | 2025-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | UNVERIFIED | OFFICIAL_API | NULL | IMF WEO series PCPIPCH (Inflation rate, average consumer prices, annual % change) for IDN. Latest completed reference year 2025 = 1.9%. Surrounding series: 2023 3.7, 2024 2.3, 2025 1.9, 2026 3 (forward years are WEO projections). Quote is the verbatim JSON fragment from the IDN object. TWO HONEST CAVEATS: (1) the endpoint IGNORES the /IDN path filter and returns all 228 economies in one blob — I parsed the 'IDN' key specifically rather than trusting the first rows returned, which belonged to SDN and AFG; a naive read of this endpoint will silently attribute Sudan's data to Indonesia. (2) I did not establish the WEO vintage or whether the 2025 figure is a final actual or a staff estimate; given the retrieval date it is likely an estimate from a 2026 WEO round. Value is well inside the 0-50 band, low is better. SOURCE QUOTE, verbatim from the page on the retrieval date: “"2023":3.7,"2024":2.3,"2025":1.9,"2026":3,"2027":2.6” | seed | seed | dbb11b67-8732-4cba-8431-cfdd8adfdecb | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 55 | b56f58be-2186-4640-a45c-90afed4436f8 | id | english-proficiency | 471 | EF EPI score (300-700) | https://www.ef.com/wwen/epi/regions/asia/indonesia/ | EF Education First | 2025-01-01 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | EF EPI 2025 edition: Indonesia score 471, global ranking #80, against a global average of 488 — i.e. below average, in EF's 'Low Proficiency' region of the scale. Skill breakdown on the same page: Reading 491, Listening 454, Writing 479. Edition year confirmed from the page's own assets: the fact sheet link is .../fact-sheets/2025/ef-epi-fact-sheet-indonesia-english.pdf and the nav carries 'Global Ranking of Countries and Regions 2025'. effectiveAt set to the start of the 2025 edition's reference year; EF does not state a precise publication date on this page, so I did not invent one. LICENSING, AS FLAGGED IN THE BRIEF: EF is the only source for this metric, and EF EPI's commercial-reuse licensing is UNRESOLVED — this observation should not be treated as cleared for redistribution until that is settled. SOURCE QUOTE, verbatim from the page on the retrieval date: “Global ranking: #80 Indonesia EF EPI score : 471 3 3pt on previous score Global average score : 488” | seed | seed | 6bc29bc1-721b-42f8-81aa-742bae617dab | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 56 | 2b4ed049-797a-459f-96b0-1e0978640d0e | id | expropriation-risk | UNRESOLVED | property-rights index (0-100, higher better) | https://www.heritage.org/index/pages/country-pages/indonesia | The Heritage Foundation | 2026-07-17 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | BLOCKED, NOT DEAD — both permitted sources refused the fetcher while remaining live. Heritage: the country page 301-redirects to https://economicfreedom.heritage.org/, and both the original URL and the new host return HTTP 403 to curl with a browser UA (3.3KB block page, no data). Fraser: https://www.fraserinstitute.org/economic-freedom/dataset?...&countries=IDN also returns HTTP 403. I did not substitute a worse source. Specifically, I did NOT fall back to Numbeo or to any of the prohibited non-commercial datasets, and I did not reach for the WGI rule-of-law figure as a stand-in — it measures a different construct and reusing it here would launder one number into two axes. DIRECTION REMINDER for whoever resolves this: this axis is a RIGHTS score, higher is better, so a low Heritage Property Rights score means WORSE protection. Resolving this likely needs a browser-based fetch (Heritage serves the data fine to real browsers) or Fraser's published dataset download. | seed | seed | ef5803a0-0609-4bc3-b126-7ce81fec3050 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 57 | a515127b-10ea-49a5-af78-cc5f5efa0913 | id | government-receptivity | 3 | rubric 0-5 | https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G | Direktorat Jenderal Imigrasi, Kementerian Imigrasi dan Pemasyarakatan, Republic of Indonesia | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 3: Named policy fit: a programme this project class demonstrably qualifies for, published eligibility criteria (SEZ regime, education-hub licence, digital-nomad statute). THE FACTS: · Indonesia operates a published remote-worker visa (E33G, 'Visa Rumah Kedua Pekerja Jarak Jauh') with objective published eligibility criteria — this is anchor 3's own enumerated example of a 'digital-nomad statute'. https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G “Rekening bank yang membuktikan penghasilan berupa gaji atau penghasilan senilai paling sedikit US$60.000 per tahun” · The KEK (Special Economic Zone) regime publishes a standing fiscal-incentive schedule with eligibility tied to a zone's designated 'kegiatan utama' (main activity). It names no project class — on its own this is anchor 2 only. https://kek.go.id/id/investment/incentive “Pemberian Tax Holiday selama 10-20 Tahun bagi investor yang melakukan penanaman modal pada kegiatan utama di KEK dan Tax Allowance untuk kegiatan di luar kegiatan utama.” · KEK Singhasari (est. PP 68/2019, Malang) is one of Indonesia's two designated Digital SEZs and is operating specifically to host world-class education providers — King's College London opened postgraduate admissions there. An education/digital-hub node is a class this named zone demonstrably qualifies for, which is what lifts the KEK regime from anchor 2 to anchor 3. https://kek.go.id/media/press/pemerintah-perluas-akses-pendidikan-kelas-dunia-di-dalam-negeri-kcl-singhasari-buka-pendaftaran-mahasiswa-baru “Pemerintah Perluas Akses Pendidikan Kelas Dunia di Dalam Negeri, KCL Singhasari Buka Pendaftaran Mahasiswa Baru” · The only on-the-record engagement between a named empowered official and the Network School project class is MALAYSIAN, not Indonesian — this is the fact that fails anchor 4 for Indonesia. https://www.malaymail.com/news/money/2026/05/14/network-school-project-in-johor-by-ex-coinbase-tech-chief-draws-govt-interest-and-scrutiny/219910 “Network School' project in Johor by ex-Coinbase tech chief draws govt interest and scrutiny” · The RUU PFII (Bali international financial centre KEK with a common-law regime and eased immigration/residency) was still under first-level DPR deliberation with ~400 DIM outstanding as of 2026-07-14, targeted for passage 2026-07-21 — i.e. after retrievedAt. It is a bill, not a statute. https://ekonomi.bisnis.com/read/20260714/9/1987916/dpr-maraton-rapat-bahas-ruu-pusat-finansial-ri-pfii-pembahasan-400-dim-berjalan-alot “DPR Maraton Rapat Bahas RUU Pusat Finansial RI (PFII), Pembahasan 400 DIM Berjalan Alot” THE FACTS: (1) Indonesia has an in-force remote-worker visa statute, E33G, with published objective criteria (US$60,000/yr income, foreign employment contract, 1-year stay) — imigrasi.go.id. (2) The KEK regime publishes a standing incentive schedule (10-20yr tax holiday for a zone's kegiatan utama) — kek.go.id. (3) KEK Singhasari (PP 68/2019) is a designated Digital SEZ operating expressly to host foreign education providers; King's College London has opened admissions there — kek.go.id press. THE MAPPING: anchor 3. Anchor 3's own enumeration is 'SEZ regime, education-hub licence, digital-nomad statute' and Indonesia has two of the three, with an education/digital node demonstrably inside KEK Singhasari's designated main activity. WHAT ARGUED FOR HIGHER AND WHY IT DID NOT REACH: (a) Anchor 4 needs a named empowered official/agency engaged with THIS project or a close analogue ON THE RECORD. I searched for it and found the opposite — the only such engagement on record is Malaysia's Digital Minister Gobind Singh Deo with Network School in Johor (malaymail, 2026-05-14). Nothing equivalent from any Indonesian official. Anchor 4 does not reach. (b) The RUU PFII — a Bali financial-centre KEK with a special common-law regime, 0% tax proposals and eased immigration/residency — reads like anchor 5 material, but as of 2026-07-17 it is a bill with ~400 DIM still being fought over, targeted for passage on 2026-07-21, four days after retrievedAt. An intention is not a statute; it caps nothing upward. If it is gazetted naming the zone's activity classes, this moves to 4 or 5. WHAT ARGUED FOR LOWER: on the Malaysia worked example's own logic, the KEK incentive schedule names no project class and would cap at 2. It reaches 3 only because KEK Singhasari is a NAMED education/digital zone whose designated kegiatan utama the class fits, and because E33G is literally anchor 3's 'digital-nomad statute'. Strip either and this is a 2. Nothing found supports 0 or 1: no named Indonesian official has opposed, restricted or investigated this project class. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS RE-FETCHED. E33G (imigrasi.go.id): both quotes verbatim — 'Rekening bank yang membuktikan penghasilan berupa gaji atau penghasilan senilai paling sedikit US$60.000 per tahun' present; purpose 'Tinggal di Indonesia untuk melaksanakan tugas dari perusahaan luar negeri' present; 1-year stay; no nationality restriction. KEK incentive (kek.go.id): the page is JS-rendered and returns blank to WebFetch, but curl recovers the embedded JSON containing the Tax Holiday 10-20 Tahun / kegiatan utama sentence verbatim. KCL Singhasari press release: headline verbatim, and the body confirms KCL runs MSc Digital Economies / MA Digital Futures postgraduate programmes at KEK Singhasari. Malaymail 403'd to WebFetch but returns 200 via curl: headline matches (actual text has a comma — 'draws govt interest, and scrutiny'), Gobind Singh Deo named, and Indonesia appears ZERO times — the negative claim that no Indonesian official is engaged holds. Bisnis.com is bot-protected (202, empty body), but the quoted headline is exactly the article title returned by search, and RUU PFII's status is independently corroborated: still level-I deliberation, ~400 DIM, targeted for 21 July 2026 plenary — a bill, not a statute. MAPPING CHECKED FOR UPWARD INFERENCE — NONE FOUND. Anchor 3 enumerates 'digital-nomad statute' in its own words, and E33G is one, in force, with published objective criteria. Anchor 3 is carried by E33G ALONE, independent of the KEK reasoning, which makes the score robust. Anchor 4 is correctly refused: it demands a named empowered official engaged with THIS project on the record, and the only such engagement is Malaysian — I verified the absence, not merely the presence. Anchor 5 is correctly refused: RUU PFII is precisely the 'intention is not a statute' trap the rubric warns about, and the note names it as such rather than banking it. Not lower than 3: anchor 2 ('generic openness only, not naming this project class') does not fit, because E33G names the activity class. ONE OVERREACH, NOT SCORE-CHANGING: fact 3 asserts KEK Singhasari is 'one of Indonesia's two designated Digital SEZs' whose designated 'kegiatan utama' the class fits — neither the phrase 'kegiatan utama' nor a Digital-SEZ designation appears in the cited press release, which only establishes that KCL runs education programmes there. That prop is weaker than stated, but the note itself says 'strip either and this is a 2', and E33G survives untouched, so the anchor holds. | seed | seed | 966d8809-018f-4173-9e4c-8d714c13c87d | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 58 | 2d274078-a555-4c12-ba85-c31329817b7e | id | visa-preference | 4 | rubric 0-5 | https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G | Direktorat Jenderal Imigrasi, Republic of Indonesia | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A preferential route exists, is in force, open AS OF RIGHT on published criteria, covers ALL FOUR cohorts. THE FACTS: · E33G is an in-force preferential long-stay route for remote workers, purpose-stated, with published objective criteria and no nationality condition stated. Stay is one year. https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G “Tinggal di Indonesia untuk melaksanakan tugas dari perusahaan luar negeri” · E33G's published criteria are financial and contractual, not nationality-based — US/EU/IN/CN applicants are all eligible on the same terms. https://www.imigrasi.go.id/wna/daftar-visa-indonesia/E33G “bukti memiliki biaya hidup selama berada di wilayah Indonesia berupa rekening koran 3 bulan terakhir” · Indonesia additionally runs a Golden Visa with published multi-year terms (5 and 10 years) under Permenkumham 22/2023 — but it is keyed to investment thresholds, not to this activity class. https://www.imigrasi.go.id/siaran_pers/2023/09/02/aturan-hukum-golden-visa-disahkan-tarik-orang-asing-berkualitas-untuk-berinvestasi?lang=en-US “Golden Visas are visas granted to foreigners as a basis for a stay permit for 5 to 10 years to support the national economy.” · Countervailing: the ordinary visa-free route was withdrawn from all four cohorts. Perpres 95/2024 replaced Perpres 21/2016, cutting visa-free entry from 169 countries to 13 (ASEAN plus Timor Leste, Suriname, Colombia, Hong Kong). US, EU, India and China are excluded and must use e-VOA or a visa. https://kediri.imigrasi.go.id/bebas-visa-kunjungan-untuk-159-negara-dihentikan-sementara-ini-alasannya/ “Bebas Visa Kunjungan untuk 159 Negara Dihentikan Sementara” THE FACTS: E33G ('Visa Rumah Kedua Pekerja Jarak Jauh') is in force, purpose-built for people performing duties for a foreign company, grants one year of stay, and is granted on published objective criteria (US$60k/yr income, 3-month bank statements, foreign employment contract) with no nationality condition — so all four cohorts are covered. The Golden Visa (Permenkumham 22/2023, promulgated 2023-08-30) adds published 5- and 10-year terms. THE MAPPING: anchor 4. WHAT ARGUED FOR 5 AND WHY IT DID NOT REACH: anchor 5 is a conjunction — designed for this activity class AND published multi-year terms. Indonesia satisfies each limb with a DIFFERENT instrument and neither instrument with both. E33G is squarely a remote-work route but its published term is one year, not multi-year. The Golden Visa has published multi-year terms but is keyed to capital thresholds (US$350k portfolio / US$2.5m company formation), not to remote work, education or talent — it is not FOR this kind of project. The rubric's own note also forbids treating nationality-agnostic as preferential-by-itself, which is exactly what E33G is. So 4, not 5. WHAT ARGUED FOR LOWER: two things. (a) Anchor 2 ('wrong activity, or a cohort excluded') is arguable — E33G requires an employment contract with a foreign company and prohibits selling goods or services in Indonesia, which excludes founders, the self-employed and anyone earning locally; if node members are not foreign-employees the route does not fit them and this drops to 2. I did not drop it because Golden Visa E28C (US$350k portfolio, 5yr) is an independent in-force preferential route with published criteria open to the same cohorts. (b) Anchor 0 ('a named restriction or exclusion touching US/EU/IN/CN') has a real claim: Perpres 95/2024 removed visa-free entry from all four cohorts. I did not score 0 because that touches the ORDINARY route, and this subject asks whether a PREFERENTIAL route exists — it does, and the four cohorts can use it. Resolving what activities node members will actually perform is what would move this to 5 or down to 2. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS RE-FETCHED. E33G both quotes verbatim on imigrasi.go.id — purpose-statement and the 3-month rekening koran requirement; requirements are financial/contractual ('Perjanjian kerja dengan perusahaan yang didirikan di luar Wilayah Indonesia'), with no nationality condition, so all four cohorts are eligible on identical terms. Golden Visa press release quote verbatim: 'Golden Visas are visas granted to foreigners as a basis for a stay permit for 5 to 10 years to support the national economy'; Permenkumham 22/2023 confirmed, thresholds confirmed (US$2.5m/US$5m company formation; US$350k/US$700k portfolio), and the page makes NO mention of remote-work or employment-based eligibility — corroborating that it is investment-keyed, not for this activity class. Kediri imigrasi headline verbatim. Perpres 95/2024 independently corroborated (promulgated 2024-08-29, revoked Perpres 21/2016, 169 countries cut to 13, and the 13-country list matches the note exactly: ASEAN + Timor Leste, Suriname, Colombia, Hong Kong — US/EU/IN/CN excluded). MAPPING CHECKED FOR UPWARD INFERENCE — NONE FOUND, AND THE HARDEST TRAP WAS AVOIDED. The rubric's note ('nationality-agnostic is NOT preferential... a route open to everyone equally is a 4 at most') is exactly the trap here, and the note names it and stops at 4 rather than reading E33G's remote-work purpose as a 5. Anchor 5 is a conjunction and Indonesia satisfies its two limbs with two DIFFERENT instruments: E33G is for the activity class but its published term is 1 year (verified: '1 year, with possibility of extension'), and the Golden Visa has published multi-year terms but is keyed to capital, not remote work. Neither instrument carries both limbs, so 5 does not reach. This is the correct and conservative read. NOT LOWER, ON TWO CHALLENGES I TESTED. (a) Anchor 0 has real surface appeal — Perpres 95/2024 IS 'a named... exclusion touching US/EU/IN/CN' and it is verified. But anchor 0's own subject is 'The ordinary route', and this axis asks whether a PREFERENTIAL route exists; a verified in-force preferential route open to all four cohorts cannot sit at an anchor that presupposes none. (b) Anchor 2 ('wrong activity... or a cohort excluded') is the live risk the note honestly flags: E33G requires foreign employment and bars earning locally, so founders and the self-employed do not fit it. 'ALL FOUR COHORTS' in anchor 4 means the four nationalities, which E33G plainly covers, so the cohort limb is met and 4 stands. The note discloses this dependency rather than hiding it. | seed | seed | 2c335d41-8a47-4d4b-8046-753d24395f3f | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 59 | a08eb069-3484-43be-9d77-882cb23da9a8 | id | web3-support | 4 | rubric 0-5 | https://www.ojk.go.id/id/regulasi/Documents/Pages/POJK-27-2024-AKD-AK/POJK%2027%20Tahun%202024%20Penyelenggaraan%20Perdagangan%20Aset%20Keuangan%20Digital%20Termasuk%20Aset%20Kripto%20AKD%20AK.pdf | Otoritas Jasa Keuangan (OJK) | 2025-01-10 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: Framework PLUS published tax treatment PLUS documented public-sector engagement. THE FACTS: · BANKING IS SOURCED, AND IT IS MANDATED, NOT MERELY TOLERATED. POJK 27/2024 Pasal 87(2) requires a licensed crypto trader's segregated customer-funds account to be opened at a commercial bank (bank umum) that OJK has approved as a consumer-fund depository bank. A regime cannot compel commercial-bank accounts for licensed crypto firms while banks decline service in practice — this is what excludes anchor 1 and satisfies anchor 3's banking limb. https://www.ojk.go.id/id/regulasi/Documents/Pages/POJK-27-2024-AKD-AK/POJK%2027%20Tahun%202024%20Penyelenggaraan%20Perdagangan%20Aset%20Keuangan%20Digital%20Termasuk%20Aset%20Kripto%20AKD%20AK.pdf “Rekening yang terpisah sebagaimana dimaksud pada ayat (1) hanya dapat dibuka pada bank umum yang menjadi bank penyimpan dana dan telah mendapat persetujuan dari Otoritas Jasa Keuangan sebagai bank penyimpan dana Konsumen.” · Customer funds move into that account by ordinary interbank transfer or licensed e-money — i.e. the licensed crypto sector runs on the regulated banking rails. https://www.ojk.go.id/id/regulasi/Documents/Pages/POJK-27-2024-AKD-AK/POJK%2027%20Tahun%202024%20Penyelenggaraan%20Perdagangan%20Aset%20Keuangan%20Digital%20Termasuk%20Aset%20Kripto%20AKD%20AK.pdf “Penempatan dana Konsumen pada rekening yang terpisah dari Pedagang sebagaimana dimaksud dalam Pasal 85 ayat (1) huruf a, dilakukan melalui pemindahbukuan antar rekening Bank atau melalui uang elektronik.” · Published tax treatment exists: PMK 50/2025, effective 2025-08-01, removes VAT on crypto transfers and sets final income tax of 0.21% of transaction value on licensed domestic platforms (1% via foreign PMSE platforms). https://www.pajak.go.id/en/node/118105 “Pada 1 Agustus 2025, lanskap perpajakan aset kripto di Indonesia memasuki babak baru.” · Documented public-sector engagement: OJK runs a named Regulatory Sandbox under POJK 3/2024 and its published participant list (as of 2025-12-15) contains four crypto/digital-asset business models, including a rupiah stablecoin. https://ojk.go.id/en/Fungsi-Utama/ITSK/Regulatory-Sandbox/Documents/List%20of%20Sandbox%20Participants%20Position%20as%20of%2015%20December%202025.pdf “PT ADHYOKA BERKAH MAJU Product name: IDRP Business Model in the Trial Process: Stablecoin S-327/IK.01/2025 (11 June 2025)” · Countervailing, and decisive against anchor 5: Indonesia does not accept digital assets as payment and enforces against it. The Governor of Bali publicly threatened enforcement against tourists paying in crypto. https://www.thestar.com.my/tech/tech-news/2023/05/29/stop-paying-in-crypto-bali-governor-warns-foreign-tourists “Foreign tourists who behave inappropriately, do activities that are not allowed in their visa permit, use crypto as a means of payment, and violate other provisions will be dealt with firmly.” BANKING, per the rubric's note, is sourced and is the load-bearing fact: POJK 27/2024 Pasal 86-87 does not merely permit banking for licensed crypto traders, it COMPELS it — segregated customer funds may only sit in an OJK-approved bank umum, and customer money enters by interbank transfer. Banks plainly do not decline service, so anchor 1 is out and anchor 3's banking limb is met. THE FACTS: (1) licensed framework — POJK 27/2024, in force 2025-01-10, moving crypto from Bappebti to OJK with licensed exchange, traders, clearing and custodians; (2) published tax treatment — PMK 50/2025 effective 2025-08-01, 0.21% final income tax, VAT removed (pajak.go.id); (3) documented public-sector engagement — the OJK Regulatory Sandbox under POJK 3/2024, with a published participant list naming four digital-asset business models. THE MAPPING: anchor 4 — framework PLUS published tax treatment PLUS documented public-sector engagement, all three present and all three primary-sourced. WHAT ARGUED FOR 5 AND WHY IT DID NOT REACH: this is the closest call in the set and a verifier could legitimately land the other way. Anchor 5's enumeration literally includes the word 'sandbox', and OJK's is a named government sandbox admitting a rupiah stablecoin (IDRP) and a crypto fund manager. Read token-by-token, that is a 5. I did not score 5 because the anchor's governing clause is 'Active public-sector adoption', and the OJK sandbox is not adoption — it is a supervisory antechamber to a private licence, a gate the state operates ON firms rather than a programme the state runs. The affirmative evidence points the other way: Bank Indonesia holds the rupiah as sole legal tender and Bali's governor has threatened deportation and criminal penalties for crypto payments, which is the negation of 'acceptance of digital assets'. A named government programme USING digital assets — a settled CBDC rollout, a state acceptance of crypto, a public procurement — would move this to 5. WHAT ARGUED FOR LOWER: 3 is what you get if you discount the sandbox and the tax rule as not amounting to 'engagement'; I rejected that because a 0.21% final rate written into a Minister of Finance regulation and a published sandbox roster with named crypto participants are both instruments, not posture. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS RE-FETCHED — AND THE BANKING LIMB, WHICH THE RUBRIC SAYS IS UNSOURCEABLE, IS GENUINELY SOURCED. I downloaded the POJK 27/2024 PDF from ojk.go.id and extracted both quotes verbatim: Pasal 87(2) 'Rekening yang terpisah sebagaimana dimaksud pada ayat (1) hanya dapat dibuka pada bank umum yang menjadi bank penyimpan dana dan telah mendapat persetujuan dari Otoritas Jasa Keuangan sebagai bank penyimpan dana Konsumen', and the Pasal 86 pemindahbukuan sentence. The inference is sound and is the strongest reasoning in this set: a regime that COMPELS licensed crypto traders to hold segregated customer funds at an OJK-approved bank umum is logically incompatible with anchor 1's 'banks decline service in practice'. This is a regulatory obligation, not a survey of bank behaviour, so it is a legitimate primary source for the limb DATA-SOURCES.md §194 calls unautomatable. PMK 50/2025 verified on pajak.go.id: opening sentence verbatim, 0.21% final rate and VAT removal both quoted on-page. OJK sandbox PDF verified verbatim: 'PT ADHYOKA BERKAH MAJU / Product name: IDRP / Business Model in the Trial Process: Stablecoin / S-327/IK.01/2025 (11 June 2025)', plus PT DANA KRIPTO INDONESIA as Crypto Fund Manager. TheStar verified: the Wayan Koster quote is present verbatim. MAPPING. All three of anchor 4's conjuncts are present and each is primary-sourced: framework (POJK 27/2024) + published tax treatment (PMK 50/2025) + documented public-sector engagement (the OJK sandbox roster). Not lower: anchor 3 would require discounting a Minister-of-Finance rate regulation and a published sandbox roster as mere posture, which they are not — both are instruments. ON ANCHOR 5, THE NOTE RESOLVED A GENUINE AMBIGUITY DOWNWARD, WHICH IS THE RIGHT DIRECTION. Anchor 5's enumeration literally contains the word 'sandbox', and OJK's is a named government sandbox admitting a rupiah stablecoin — read token-by-token that is a 5, and the note says so openly rather than concealing the tension. It stops at 4 because the anchor's governing clause is 'Active public-sector ADOPTION', and a supervisory gate the state operates ON private firms is not the state adopting anything; the affirmative record points the opposite way, with the verified Koster threat of enforcement against crypto payments being the negation of 'acceptance of digital assets'. Since my mandate is to catch inference UPWARD and to default downward when the anchor is uncertain, a defensible reading that declines the higher anchor is confirmed, not remapped. | seed | seed | 81dce910-dee0-4149-87fd-750fe5d13951 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 60 | 9210552f-3980-40c1-b175-02608d542bd3 | id | capital-controls | 1 | rubric 0-5 | https://jdih.kemenkeu.go.id/api/download/fulltext/2007/25TAHUN2007UU.htm | Jaringan Dokumentasi dan Informasi Hukum, Kementerian Keuangan (Ministry of Finance JDIH), Republic of Indonesia | 2007-04-26 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 1: Reporting or registration only; no approval needed; no documented delays. THE FACTS: · Law 25/2007 on Investment, Pasal 8(3), grants investors a statutory right to transfer and repatriate in foreign currency — capital, profits, bank interest, dividends, other income, loan repayments, royalties, foreign staff earnings, liquidation and sale proceeds, and compensation. https://jdih.kemenkeu.go.id/api/download/fulltext/2007/25TAHUN2007UU.htm “Penanam modal diberi hak untuk melakukan transfer dan repatriasi dalam valuta asing, antara lain terhadap: a. modal; b. keuntungan, bunga bank, deviden, dan pendapatan lain;” · THIS IS THE ANCHOR-1 FACT. Pasal 8(5)(a) preserves to government only the power to require REPORTING of fund transfers — not to approve them. The statute reserves a reporting power, not an approval power. https://jdih.kemenkeu.go.id/api/download/fulltext/2007/25TAHUN2007UU.htm “kewenangan Pemerintah untuk memberlakukan ketentuan peraturan perundang-undangan yang mewajibkan pelaporan pelaksanaan transfer dana” · The right is exercised subject to subordinate regulation — the hook through which Bank Indonesia's underlying-document thresholds operate. https://jdih.kemenkeu.go.id/api/download/fulltext/2007/25TAHUN2007UU.htm “Hak untuk melakukan transfer dan repatriasi sebagaimana dimaksud pada ayat (3) dilakukan sesuai dengan ketentuan peraturan perundang-undangan.” · Bank Indonesia does adjust the thresholds above which FX-against-rupiah transactions require underlying-transaction documents; PADG 7/2026 (26 March 2026, effective 1 April 2026) is the current amendment. This is a documentation requirement discharged by producing the underlying transaction, not a discretionary approval. https://www.bi.go.id/id/publikasi/peraturan/Pages/PADG_072026.aspx “Peraturan Anggota Dewan Gubernur Nomor 7 Tahun 2026” · Countervailing: PP 8/2025 (effective 2025-03-01) requires 100% of natural-resource export proceeds to be retained in the domestic financial system for 12 months. This is a genuine quantitative retention rule — but it binds mining, plantation, forestry and fisheries exporters, not a node. https://jdih.kemenkeu.go.id/dok/pp-8-tahun-2025 “PP 8 TAHUN 2025” SOURCE DISCIPLINE, AS THE RUBRIC DEMANDS I SAY SO EXPLICITLY: I did NOT resolve this from IMF AREAER. AREAER online (elibrary-areaer.imf.org) redirects to an IMF bookstore auth gateway and the Indonesia 2025 Article IV staff report XML (elibrary.imf.org/view/journals/002/2026/010/) and the Board press release (imf.org PR-26010) both returned HTTP 403. Blocks reported, not worked around. I reasoned instead from the primary statute plus Bank Indonesia's own regulation, and the score should be read with that discount. THE FACTS: Law 25/2007 Pasal 8(3) grants a statutory repatriation right over capital, profits, dividends, loan repayments, royalties, foreign staff earnings and liquidation proceeds; Pasal 8(5)(a) reserves to government a power to require REPORTING of transfers — conspicuously not a power to approve them; Pasal 8(4) subordinates the right to regulation, under which BI sets thresholds above which underlying-transaction documents must be produced (PADG 7/2026, effective 2026-04-01). THE MAPPING: anchor 1 — reporting/documentation only, no approval, and I found no documented delays. WHAT ARGUED FOR HIGHER AND WHY IT DID NOT REACH: three things. (a) BI has been ratcheting the underlying-document threshold down hard through 2026 (reported: US$100k to US$50k in March, to US$25k, to US$15k in June, and a US$10k cash limit from 1 July 2026). I could not extract the numeric threshold from bi.go.id itself — the regulation page gives the instrument and dates but not the figure — so I am not scoring on it. Even taken at face value it is documentation, not approval: an entity with a real underlying transaction transacts above the threshold. That is anchor 1, not anchor 2's 'approval required... routinely granted'. (b) PP 8/2025's 100%/12-month DHE SDA retention IS an anchor-4-shaped quantitative limit, and the IMF and the Indonesian authorities are on record disagreeing about whether it is a capital flow measure at all. It does not reach here because it binds natural-resource exporters and does not touch a node's normal operation. (c) Anchor 0 (no controls at all) fails because the reporting obligation and the underlying-document regime are real. WHAT WOULD MOVE THIS: the AREAER Indonesia chapter, if obtainable, is decisive either way — if it records Indonesia as maintaining approval requirements or documented delays on transfers, this is a 2 or 3. Treat this score as CORROBORATED, not verified against the named source. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS RE-FETCHED. All three Law 25/2007 quotes are verbatim in the kemenkeu JDIH fulltext. Critically, the anchor-1 fact survives inspection of its surrounding text: Pasal 8(5) opens 'Ketentuan sebagaimana dimaksud pada ayat (1) tidak mengurangi:' and then (a) reserves only 'kewenangan Pemerintah untuk memberlakukan ketentuan peraturan perundang-undangan yang mewajibkan pelaporan pelaksanaan transfer dana' — a REPORTING power. I checked the adjacent subsections (b) tax/royalty rights, (c) creditor-protection law, (d) other law; none reserves an approval power over transfers. The note's central distinction therefore holds on the face of the statute rather than by characterisation. PADG 7/2026 verified on bi.go.id: it amends PADG 11/2024 on FX market transactions, effective 1 April 2026, and adjusts thresholds above which underlying-transaction DOCUMENTS must be provided — the page describes no approval mechanism. PP 8/2025 verified: its JDIH title confirms it amends PP 36/2023 on 'Devisa Hasil Ekspor dari Kegiatan Pengusahaan, Pengelolaan, dan/atau Pengolahan Sumber Daya Alam', which corroborates that the 100%/12-month retention binds natural-resource exporters and not a node. SOURCE DISCIPLINE SATISFIED ON THE RUBRIC'S OWN TERMS. The rubric names IMF AREAER and requires that reasoning from a central bank's posture instead 'MUST say so'. The note says so in its first line, names the exact failures (AREAER redirects to an auth gateway; the Article IV XML and PR-26010 both 403), reports the blocks rather than working around them, and instructs the reader to treat the score as corroborated rather than verified. That is compliance with the discipline, not evasion of it. MAPPING — AND THIS AXIS IS REVERSE-SCORED, SO I TESTED FOR INFERENCE DOWNWARD (THE FAVOURABLE DIRECTION). Anchor 1 is 'reporting or registration only; no approval needed'. An underlying-document threshold is discharged by producing a real underlying transaction; it does not vest discretion in an authority, so it is not anchor 2's 'approval required... routinely granted on published criteria'. Anchor 0 correctly fails because the reporting obligation is real. The note also declines to score on the reported threshold ratchet ($100k→$50k→$25k→$15k) precisely because it could not extract the figures from bi.go.id — refusing to score on an unverified number is the correct discipline. The residual risk is stated by the note itself and I endorse it: if the AREAER Indonesia chapter records approval requirements or documented delays, this becomes a 2 or 3. | seed | seed | e62a67a6-fc0c-43c7-b6e6-43b825cad246 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 61 | d9825f96-61d0-46b9-9f89-50fbf8022dc8 | id | policy-reversal-precedent | 4 | rubric 0-5 | https://kediri.imigrasi.go.id/bebas-visa-kunjungan-untuk-159-negara-dihentikan-sementara-ini-alasannya/ | Kantor Imigrasi Kelas I TPI Kediri, Direktorat Jenderal Imigrasi, Republic of Indonesia | 2023-06-07 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A programme closed or suspended to new entrants with little notice. THE FACTS: · THE CITED INSTANCE. Indonesia granted visa-free entry to 169 countries under Perpres 21/2016, then suspended it for 159 of them by Ministerial Decree M.HH-GR.01.07 of 2023, signed by Minister of Law and Human Rights Yasonna Laoly on 7 June 2023 and effective the same day. Ten ASEAN-region states were left. No transition period is stated. https://kediri.imigrasi.go.id/bebas-visa-kunjungan-untuk-159-negara-dihentikan-sementara-ini-alasannya/ “Bebas Visa Kunjungan untuk 159 Negara Dihentikan Sementara, Ini Alasannya!” · The stated grounds were prospective and generic — public-order disturbance and disease transmission — not any conduct by the affected states, i.e. this was a policy reversal, not enforcement against a defaulting beneficiary. https://kediri.imigrasi.go.id/bebas-visa-kunjungan-untuk-159-negara-dihentikan-sementara-ini-alasannya/ “perturbation of public order” · The 'temporary' suspension was then made permanent and formalised: Perpres 95/2024 (effective 2024-08-29) revoked and replaced Perpres 21/2016, re-granting visa-free entry to only 13 countries. The 2016 regime as granted no longer exists. https://pro.hukumonline.com/a/lt66da88ae23a6b/pembebasan-visa-kunjungan--hanya-13-negara-yang-kini-memenuhi-syarat “Pembebasan Visa Kunjungan: Hanya 13 Negara yang Kini Memenuhi Syarat” · Argued for 5: on 6 January 2022 the government revoked 2,078 mineral and coal mining permits, 192 forestry permits and 34,448 ha of cultivation rights (HGU) from existing holders. Officially predicated on holders' non-performance — no work plans, abandoned land, misuse or transfer of permits. https://en.tempo.co/read/1547186/jokowi-revokes-2078-mineral-coal-mining-permits-today “Jokowi Revokes 2,078 Mineral, Coal Mining Permits Today” THE FACTS: Indonesia granted visa-free entry to 169 countries (Perpres 21/2016), then Minister of Law and Human Rights Yasonna Laoly suspended it for 159 of them by Kepmenkumham M.HH-GR.01.07, signed 7 June 2023 and effective 7 June 2023 — same-day, no transition period, on generic prospective grounds (public order, disease) rather than anything the beneficiaries did. Perpres 95/2024 then revoked Perpres 21/2016 outright and re-granted visa-free entry to 13 countries only. THE MAPPING: anchor 4 — 'a programme closed or suspended to new entrants with little notice'. Same-day effect on a benefit regime nine years old is the paradigm case. WHAT ARGUED FOR 5 AND WHY IT DID NOT REACH: two things. (a) Perpres 95/2024 arguably 'abolished a regime without transition' — 169 countries to 13 is closer to abolition than amendment. I did not reach 5 because the regime was not abolished; it was reconstituted at a fraction of its scope, and visa-free entry has no 'existing holders' to revoke from — nobody holds a grant. (b) The January 2022 mass revocation — 2,078 mining permits, 192 forestry permits and 34,448 ha of HGU land rights stripped from existing holders — is anchor 5's own language ('grants revoked from existing holders') and it is the single most alarming item in Indonesia's record, especially given that anchor 5 of government-receptivity contemplates ALLOCATED LAND as the instrument this project class would rely on. I did not score 5 on it for two reasons, and a verifier should weigh both: first, the revocations were officially predicated on the holders' own non-compliance (no work plan filed, land abandoned, permits misused or transferred), which is enforcement of licence conditions against defaulters rather than reversal of a grant in good standing; second, I could not fetch a primary source — setkab.go.id serves a JS challenge and its TLS chain fails verification from here, so the fact rests on Tempo (reputable media), not the Cabinet Secretariat. Blocks reported. If a primary source shows compliant holders were swept up, this is a 5. WHAT ARGUED FOR LOWER: anchors 2 and 3 (tightened prospectively, holders grandfathered / with transition) do not fit — there was no grandfathering and no transition. Anchors 0 and 1 are unavailable on their face: a reversal is on record within the last 10 years, so 'no reversal found' is false. Note the direct relevance to this project class: the withdrawn benefit is exactly the ordinary-entry route for the US/EU/IN/CN cohorts. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS RE-FETCHED, WITH TWO DEFECTS THAT DO NOT REACH THE ANCHOR. Verified: the kediri imigrasi headline is verbatim; Kepmenkumham M.HH-GR.01.07 of 2023, signed by Yasonna Laoly on 7 June 2023, suspending visa-free entry for 159 countries with ~10 ASEAN states left. Perpres 95/2024 is independently corroborated despite the cited source failing (see below) — promulgated 2024-08-29, revoked Perpres 21/2016, 169 countries cut to 13, exact country list matching. Tempo verified verbatim and in full: 6 January 2022, 2,078 mining permits, 192 forestry permits over 3,126,439 ha, 34,448 ha of HGU — and, decisively for the note's own reasoning, the stated grounds check out as non-compliance: 'the companies never submitted their work plan', 'revoked because they were inactive, did not make a work plan and [the lands] were neglected'. DEFECT 1 — A QUOTE THAT IS NOT ON THE PAGE. Fact 2's quote 'perturbation of public order' appears ZERO times on the kediri page; it is an English translation. The page carries 'gangguan ketertiban umum' plus disease-spread grounds, so the CLAIM (generic prospective grounds, not beneficiary conduct) is established — but the quote as written would not survive a literal re-fetch and should be replaced with the Indonesian. DEFECT 2 — hukumonline is behind a Cloudflare interstitial (403, 'Just a moment...'), so I could not check that quote. I did not treat this as fatal because I independently confirmed every element of its claim from jdih.setkab.go.id and other sources; nothing in the score rests on an unverifiable assertion. DEFECT 3 — the note says the decree was 'effective the same day'; the cited page establishes signature on 7 June 2023 and announcement on 16 June 2023, and does not state same-day effect. Overstated, but immaterial: a benefit regime withdrawn with the public told after the fact is 'little notice' either way. MAPPING. Anchor 4 — 'a programme closed or suspended to new entrants with little notice' — fits a nine-year-old visa-free regime suspended for 159 countries with no transition stated, and the rubric's stem expressly puts a 'visa regime' in scope. Anchors 2 and 3 fail for want of any grandfathering or transition; 0 and 1 are unavailable because a reversal is on record within 10 years. Anchor 5 is correctly refused on both candidates: visa-free entry has no 'existing holders' to revoke from, and the 2022 mass revocation — which I verified — was predicated on holders' own default, making it enforcement against defaulters rather than reversal of a grant in good standing. Note this axis is reverse-scored, so 4 is the harsh call, not a flattering one; the note declines to inflate it to 5 on the most alarming item in the record and states plainly what would change its mind. | seed | seed | 5b4c8319-fc7d-46ab-912a-488e03378233 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 62 | 55a7da10-2acf-4d54-b252-4a41c3beed9b | kz | visa-access | {"route":"Visa-free entry (30-day unilateral visa-free regime, Rules of entry and stay of immigrants, para. 17)","permits":["VISIT"],"maxStayDays":30,"passportSet":"US"} | NULL | https://adilet.zan.kz/eng/docs/P1200000148 | Government of the Republic of Kazakhstan — Resolution No. 148 of 21 January 2012 (official English translation, Adilet Legal Information System, Ministry of Justice) | 2022-07-07 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | NON-QUALIFYING VISIT ROUTE — recorded honestly per BRIEF.md §6 (tourist access is not permission to work, study or reside). Fails the scenario on BOTH duration (30 < 180) and activity (no WORK_REMOTE, no STUDY, no long-term RESIDE). SOURCE QUOTE, verbatim from Resolution No. 148, para. 17 (first part): "Citizens of the Commonwealth of Australia, United States of America, Republic of Austria ... [long enumerated list] ... shall enter and leave the Republic of Kazakhstan without a visa if their period of stay in the Republic of Kazakhstan does not exceed thirty calendar days after crossing the state border of the Republic of Kazakhstan, for a total of ninety calendar days within each period of one hundred and eighty calendar days." maxStayDays = 30 is the PER-ENTRY stay stated on the page; the 90-day figure is a cumulative cap within 180 days, not a single-route stay, and is not recorded as maxStayDays. WHAT IT DOES NOT PERMIT: labour of any kind is barred to visa-free entrants — Resolution 148 para. 7, verbatim: "Immigrants temporarily staying in the Republic of Kazakhstan shall be prohibited from engaging in labour activities without relevant permits issued by the local executive authority, unless otherwise stipulated by the laws of the Republic of Kazakhstan on migration and/or international treaties ratified by the Republic of Kazakhstan." No STUDY enrolment is granted by visa-free entry (education runs through the C9 education visa / a temporary residence permit per para. 10). The USA is named EXPRESSLY in para. 17, so this row is keyed 'US' on an affirmative nationality statement (not the weaker '*' basis). TRANSLATION CAVEAT: Adilet labels the English text an 'Unofficial translation' of the government instrument; it is nonetheless the authority's own consolidated legal text. Para. 17 last reworded by Decree No. 342 of 02.06.2020 and amended by Decree No. 464 of 07.07.2022 (effectiveAt reflects the latter). | seed | seed | f15b4f6a-64e3-4d8b-8807-087f87a3cdcf | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 63 | d8f711dd-bd49-4cfd-b649-0ce398da0cc4 | uz-tashkent-city | international-air-service | 1 | service (0/1) | https://ourairports.com/airports/UTTT/ | OurAirports (community open-data airport registry) | 2026-03-13 | 2026-07-20T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | D-047: the binary D-044 actually licensed, recorded on the criterion that can carry it. This row does NOT assert a route count — `flight-connectivity` stays empty for Tashkent because no route count was obtainable. VERIFIER: CONFIRMED on value (evidence/tashkent-city/verifier-verdicts.md §5); the shape defect it flagged portfolio-wide is what D-047 fixes. VERBATIM from the OurAirports facility-data panel (fetch_page, HTTP 200, 2026-07-20, every field re-fetched and CONFIRMED): 'Name: Tashkent International Airport', 'Location: Tashkent, Tashkent Region, UZBEKISTAN', 'IATA code: TAS', 'ICAO code: UZTT', 'Facility type: large_airport', 'Airline service?: yes', 'Coordinates: 41.257900,69.281197', 'Field elevation: 1,417 ft / 432 m MSL', 'Last updated: 2026-03-13 17:32:26'. 'Airline service? yes' is OurAirports' rendering of the CSV column scheduled_service="yes"; 'Facility type: large_airport' is the CSV type column. ICAO CODE CHANGE, MATERIAL FOR ANY RE-CHECK: the record's ICAO is now 'UZTT', NOT the legacy 'UTTT' — a site comment reads verbatim 'change ICAO code prefix from UT to UZ from 2025-10-02' (@espinielli, Mon, 29 Sep 2025), and the legacy code survives only under 'Keywords: UTTT'. The /UTTT/ URL still resolves with no redirect (both agents), but a checker re-deriving the URL from the current ICAO code would miss the record and could wrongly report an absence. TIER NOTE: OurAirports is a community-maintained public-domain registry, not a civil-aviation-authority primary (that would be Uzbekistan's Civil Aviation Agency / AIP). effectiveAt = the record's own 'Last updated' date — the freshest of the four cities. GEOMETRY NOTE: the airport reference point 41.257900,69.281197 is close to but distinct from the seeded city centroid [69.2401, 41.2995] (~5 km); that is geometry, not a proposed centroid change, and it was NOT converted into an `airport-travel-time` in minutes — see the held row's wall. | seed | seed | 636abe56-b863-419c-bec6-721fa143485b | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 64 | 409e9c54-268e-45c4-bf2b-0f96dd27d4af | kz | visa-access | {"route":"Visa-free entry (14-day regime for India/China/Iran, Rules of entry and stay of immigrants, para. 17)","permits":["VISIT"],"maxStayDays":14,"passportSet":"IN"} | NULL | https://adilet.zan.kz/eng/docs/P1200000148 | Government of the Republic of Kazakhstan — Resolution No. 148 of 21 January 2012 (official English translation, Adilet Legal Information System, Ministry of Justice) | 2022-07-07 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | NON-QUALIFYING VISIT ROUTE. Fails the scenario on duration (14 < 180) and activity (no WORK_REMOTE, no STUDY, no long-term RESIDE). SOURCE QUOTE, verbatim from Resolution No. 148, para. 17 (second part): "Citizens of the Islamic Republic of Iran, the People's Republic of China and the Republic of India shall enter and leave the Republic of Kazakhstan without a visa if their period of stay in the Republic of Kazakhstan does not exceed fourteen calendar days after crossing the state border of the Republic of Kazakhstan, forty-two calendar days in total during each period of one hundred and eighty calendar days." maxStayDays = 14 is the per-entry stay; 42/180 is a cumulative cap, not recorded as maxStayDays. India is named EXPRESSLY, so keyed 'IN' on an affirmative statement. Labour barred (para. 7, quoted in the US row). No STUDY enrolment granted. NOTE ON INDIA VS CHINA: unlike China (see CN row), I found NO Kazakhstan–India bilateral treaty extending India beyond the statutory 14 days; the 14-day figure is India's current per-entry stay. Para. 17 second part introduced/reworded by Decree No. 464 of 07.07.2022 (effectiveAt). | seed | seed | 129f6250-0654-42ce-8dd1-896f6f388a39 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 65 | fcd3f9e3-eba2-4b52-8b9c-b12f5bad4c81 | kz | visa-access | {"route":"Visa-free entry (statutory 14-day regime, Rules of entry and stay of immigrants, para. 17)","permits":["VISIT"],"maxStayDays":14,"passportSet":"CN"} | NULL | https://adilet.zan.kz/eng/docs/P1200000148 | Government of the Republic of Kazakhstan — Resolution No. 148 of 21 January 2012 (official English translation, Adilet Legal Information System, Ministry of Justice) | 2022-07-07 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | NON-QUALIFYING VISIT ROUTE. Fails the scenario on duration and activity regardless of the exact day count. maxStayDays = 14 is the figure stated on THIS cited page (Resolution 148 para. 17 second part, quoted in full in the IN row: China is named alongside Iran and India for a 14-day / 42-per-180 visa-free stay). IMPORTANT SUPERSESSION, recorded but NOT folded into the number because it rests on a source I did not retrieve as primary: a Kazakhstan–China bilateral mutual visa-exemption Agreement (signed 17 May 2023, Xi'an) entered into force on 10 November 2023 and, per Resolution 148 para. 4 ('If an international treaty ratified by the Republic of Kazakhstan establishes the rules other than those contained in this Code, then the rules of the international treaty will apply') and para. 6 (treaty terms govern), extends Chinese ORDINARY-passport per-entry visa-free stay to 30 calendar days (90 within 180). Secondary confirmation: The Astana Times, 'Kazakhstan – China Visa-Free Travel Comes into Effect' (https://astanatimes.com/2023/11/kazakhstan-china-visa-free-travel-comes-into-effect-intends-to-boost-tourism-ties/), which also states the agreement 'does not grant permission for work, study, or missionary work.' I kept maxStayDays at 14 (matched to the cited primary page) rather than 30, because the D-032 rule is that the number must appear on the cited source; the treaty text itself (adilet, ratification law) was not retrieved this pass. Either way the route is VISIT-only and non-qualifying. Labour barred (para. 7). No STUDY granted. | seed | seed | eb2dcc01-cf9a-4433-aee3-0f59cda72102 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 66 | 4199c000-fc06-4fde-a405-f52351bc671d | kz | visa-access | {"route":"Neo Nomad Visa (B12-1) — tourist-visa subcategory for remote workers","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":365,"passportSet":"*"} | NULL | https://adilet.zan.kz/eng/docs/V1600014531 | Ministry of Foreign Affairs & Ministry of Internal Affairs of the Republic of Kazakhstan — Rules for execution of invitations and issuance/extension of Kazakhstan visas (order reg. No. V1600014531; official English translation, Adilet LIS, Ministry of Justice) | 2024-11-10 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | THE NEAR-MISS ROUTE, and the reason the gate does not pass for any cohort: the Neo Nomad Visa permits RESIDE + WORK_REMOTE for up to a year (>= 180) but does NOT permit STUDY. SOURCE QUOTE, verbatim from the visa-category table of the Rules (row 'В12-1'): "В12-1 NeoNomad Visa | individuals who work remotely and have permanent income from foreign sources. | multiple | Up to 1 year | for the entire period of validity of the visa | A visa is issued by foreign institutions of the Republic of Kazakhstan on the basis of an invitation, if available: 1) a bank statement for the past 6 months confirming a stable monthly income of over $3,000; 2) a tax return issued by the competent authority of the country of nat[ionality]..." maxStayDays = 365: the visa is 'Up to 1 year' multiple-entry and the permitted stay is 'for the entire period of validity of the visa' (i.e. the stay is not clipped to 30/90 days as the ordinary B12 tourist visa is — the ordinary B12 row reads 'single entry visa | up to 90 calendar days | no more than 30 calendar days'). I record 365 as the first-issue maximum ('Up to 1 year'). The in-country EXTENSION is EXCLUDED per the per-entry-stay rule: the Rules provide the B12-1 'multiple entry up to 1 year - based on the applicant's petition for a period of no more than 1 (one) year' extension is granted by the territorial internal-affairs (police) body on petition — discretionary, so the further year is not summed. WORK_REMOTE is the route's EXPRESS purpose ('individuals who work remotely and have permanent income from foreign sources'); it authorises remote work for a FOREIGN employer/source, not local Kazakhstani employment. WHAT IT DOES NOT PERMIT — STUDY: the category purpose is confined to remote work with foreign income; nothing in the row or the Rules grants a B12-1 holder enrolment in an educational organisation (education is a separate track — C9 education visa, or a temporary residence permit issued to full-time students under Resolution 148 para. 10). Silence is not permission, so STUDY is NOT recorded, and the gate therefore has no single route covering all three intended activities. PASSPORT-SET BASIS FOR '*': the WEAKER form, stated explicitly per PASSPORT.md §1 — the category is defined by remote-work + foreign-income conditions with NO nationality condition, and no nationality is barred; however I did NOT find an affirmative government statement of nationality-neutrality. Nuance: the visa is issued 'on the basis of an invitation', but nationals of ~48 economically developed / migration-stable countries are exempt from the invitation requirement (per gov.kz reporting); countries off that list still qualify but must supply an invitation — so the invitation is a procedural condition, not a nationality bar. effectiveAt 2024-11-10 is the reported commencement of the B12-1 category (joint MFA/MIA order of 31 Oct 2024, effective 10 calendar days after first official publication, per secondary reporting incl. Fragomen/Vialto/Erickson Immigration Group); the precise gazette date was not retrieved, but the primary consolidated Rules retrieved 2026-07-19 confirm B12-1 is currently in force. TRANSLATION CAVEAT: Adilet labels the English text an 'Unofficial translation'. | seed | seed | 10708e37-4437-4400-be65-5b3fafc32230 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 67 | 0af74db3-6462-4444-91ba-33c53e4e152a | kz | operating-licence | true | NULL | https://adilet.zan.kz/eng/docs/Z070000319_ | Law of the Republic of Kazakhstan No. 319-III of 27 July 2007 "On Education" (consolidated), Republic of Kazakhstan — official legal information system adilet.zan.kz, Ministry of Justice | 2007-07-27 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | DOCUMENT_EXTRACTION | NULL | A FOREIGN-controlled entity MAY hold the licence to conduct educational activity in Kazakhstan. This is the OPPOSITE of Thailand's Private School Act, which imposes an express Thai-nationality / Thai-majority-shareholding bar on the licensee (PSA s.21(1), s.22(2), applied to non-formal schools by s.127). Kazakhstan's education-licensing statute contains NO nationality or ownership condition on the licence holder, and Kazakhstan law affirmatively recognises and regulates foreign educational institutions and their branches operating on its territory. CHAIN OF REASONING, from the primary instrument (Law 'On Education' No. 319-III of 27.07.2007, as amended): (1) NO NATIONALITY/OWNERSHIP BAR ON THE LICENSEE. Article 57 ('Licensing of activity in the sphere of education'), paragraph 1, in its current wording (heading reworded by Law of 13.11.2015 No. 398-V), makes the licensee simply a 'legal person' with no further nationality or ownership qualification. Article 57 verbatim (Russian): “1. Образовательная деятельность юридических лиц (далее – лицензиат) подлежит лицензированию в соответствии с настоящим Законом и законодательством Республики Казахстан о разрешениях и уведомлениях.” (≈ 'The educational activity of legal persons (hereinafter – the licensee) is subject to licensing in accordance with this Law and the legislation of the Republic of Kazakhstan on permits and notifications.'). The eligibility to hold the licence turns on qualification requirements for educational activity, not on the nationality of the owners. (2) EDUCATIONAL ORGANISATIONS OPERATE REGARDLESS OF FORM OF OWNERSHIP. The Law repeatedly regulates educational organisations 'регардлесс оф тхеир форм оф овнершип' — verbatim from the adilet English text of the Law read on the retrieval date: '...educational organizations, regardless of their form of ownership and departmental subordination, implementing general educational programs...'. Private (including foreign-owned) legal persons are squarely contemplated. (3) FOREIGN EDUCATIONAL ORGANISATIONS AND THEIR BRANCHES ARE EXPRESSLY RECOGNISED. The Law's own definitions (Article 1) refer, verbatim from the adilet English text, to 'documents on education issued by foreign educational organizations, including their branches, as well as research centers and laboratories', and define (item 57-1) 'international school – educational organization, implementing self-developed integrated educational programs, passing authorization in Organization of International Baccalaureate or international institutional accreditation'. Article 57(4-1) verbatim (Russian): “Филиалы некоммерческих организаций, имеющие статус международных школ, вправе осуществлять образовательную деятельность, подлежащую лицензированию, на основании лицензии и (или) приложения к лицензии на занятие образовательной деятельностью, выданной этой некоммерческой организации.” (≈ 'Branches of non-commercial organisations having the status of international schools are entitled to conduct educational activity subject to licensing on the basis of the licence and/or annex to the licence for educational activity issued to that non-commercial organisation.'). (4) A DEDICATED FOREIGN-INSTITUTION REGIME EXISTS AND RUNS ON THE SAME EDUCATION LICENCE. On 16 April 2026 the Minister of Science and Higher Education approved the 'Rules for the activity of international and foreign educational institutions in the Republic of Kazakhstan and their branches, and forms of their cooperation' (effective 2 May 2026), which govern foreign institutions/branches operating in Kazakhstan and expressly tie their continued operation to holding an education licence (their activity is terminated inter alia upon 'the withdrawal and deprivation of the licence or the annex to the licence for educational activity'). This is affirmative, in-force enablement of foreign educational institutions and their branches — the strong-form contrary of a nationality bar. (Reported via zakon.kz / mail.kz on 22 Apr 2026; the primary ministerial order text was NOT fetched — see fetch walls; a lawyer should pull the primary order.) This regime is why UK (Coventry, Cardiff, De Montfort, Heriot-Watt), Russian (MGIMO), and other foreign university branch campuses lawfully operate in Astana/Almaty. WHY 'true' AND NOT 'null': the licensing regime was actually located and read; the operative licensing provision names only a 'legal person' with no nationality/ownership condition, the Law affirmatively defines and admits foreign educational organisations and their branches, and a dedicated in-force ministerial regime governs foreign institutions operating under the education licence. The absence of a Thailand-style bar is therefore not mere silence — it is accompanied by affirmative enabling text, which is the strong basis for recording foreign eligibility. OPEN QUESTIONS a lawyer must close (recorded, not guessed): (a) Under the CURRENT wording the licensee is a 'legal person' (юридическое лицо); a foreign NATURAL person cannot hold the education licence directly — nor can any natural person — so foreign eligibility runs through a Kazakhstan legal person (which may be foreign-owned) or a foreign institution's branch, not through a foreign individual applicant. (b) I did not exhaustively read the qualification-requirements orders (e.g. Order V2200030721 for organisations other than higher education, and the HE qualification requirements) to rule out an ownership condition buried there; however, the fact that foreign-controlled branch campuses in fact operate under licence is strong evidence no such bar exists. (c) The AIFC (Astana International Financial Centre) education carve-out was not separately sourced; it would, if anything, be MORE permissive, so it does not disturb the 'true' finding. FETCH WALLS (fetch facts, NOT missing legal facts, per D-043): the adilet.zan.kz consolidated Law renders as a very long single document; both the English (https://adilet.zan.kz/eng/docs/Z070000319_) and Russian (https://adilet.zan.kz/rus/docs/Z070000319_) fetches returned HTTP 200 but the rendered text truncated after the early articles / the Article 1 glossary (the ~101k-character rendered-text cap was reached before the operative Article 57 body). The definition-level quotes above were read directly from the adilet English primary page; the operative Article 57 paragraphs 1 and 4-1 were therefore read from the kodeksy-kz.com legislation mirror (https://kodeksy-kz.com/ka/ob_obrazovanii/57.htm, marked 'Действующий... Проверено 10.07.2026'), a statute-republishing portal (secondary, NOT a law firm marketing page), which reproduces the official Article 57 including its full amendment footnote chain through Law of 30.12.2024 No. 148-VIII. To fully close this on primary text, fetch adilet Article 57 in a browser session that renders the full document, and fetch the 16 Apr 2026 ministerial Rules primary order. | seed | seed | 23a90432-8c1a-46f2-a765-ab3c371406a3 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 68 | 127a6ae5-baf3-480a-b983-683d424f94ca | my | lawfulness | UNRESOLVED | NULL | https://www.thestar.com.my/news/nation/2026/07/14/johor-won039t-allow-state-to-be-base-for-ideologies-against-the-law-says-mb | The Star (reporting Johor Menteri Besar Onn Hafiz Ghazi) | 2026-07-14 | 2026-07-17T00:00:00.000Z | REPUTABLE_MEDIA | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | BETTER EVIDENCE THAN THE SEED — and it moves the wrong way. This is not a generic 'no analyst review yet' UNRESOLVED; lawfulness is actively contested by the state government as of three days ago. On 2026-07-14 the Johor Menteri Besar ordered a full probe into Network School at Forest City and asked the Higher Education Ministry, Digital Ministry and regulatory agencies to determine whether its programmes require registration or approval under Malaysian law — i.e. the Malaysian state itself does not currently assert that this operation is lawful, and has said it will take 'firm action without compromise' on any breach of law, misuse of passes, or violation of licence or land conditions. Corroborated by Free Malaysia Today (https://www.freemalaysiatoday.com/category/nation/2026/07/14/network-school-in-forest-city-faces-home-ministry-probe), which reports a Home Ministry / immigration / police investigation into allegations that Israeli nationals participated using second-country passports. Note the interaction with the DE Rantau route: MDEC states the pass is open to all nationalities EXCEPT Israel. Per BRIEF.md this stays UNRESOLVED — it needs qualified Malaysian legal review, not public-data inference — but the analyst should treat the pending probe as a live downside, not a formality. Publisher is media because the underlying instrument is a ministerial direction, not a gazetted document I could fetch. SOURCE QUOTE, verbatim from the page on the retrieval date: “Johor will not allow any parties to use the state as a base to promote ideologies or movements that are against the law” | seed | seed | de6f08ba-860f-4156-ad26-d5d64d4522a5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 69 | 6cc49d70-dd63-48a1-869c-ade5cf8452ef | my | visa-access | UNRESOLVED | NULL | https://www.mdec.my/derantau | Malaysia Digital Economy Corporation (MDEC), Ministry of Digital | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed {"US": [{"route": "DE Rantau Nomad Pass (Professional Visit Pass / Pas Lawatan Ikhtisas)", "permits": ["RESIDE", "WORK_REMOTE"], "maxStayDays": 365}, {"route": and an independent agent re-fetched the source and rejected it. VERIFIER: The quote is literally on mdec.my/derantau, and it does establish exactly one row of the matrix: DE Rantau = Professional Visit Pass, stay and work up to 12 months (365 days, RESIDE + WORK_REMOTE). It establishes nothing else. I asked the page directly whether it mentions Student Pass, MM2H, or the China visa exemption and it does not. Yet the claimed value asserts Student Pass (365d, RESIDE+STUDY), MM2H Social Visit Pass (1825d, RESIDE) for all four nationalities, and a China-Malaysia mutual visa exemption (30d) — three of the four routes, including the oddly precise 1825-day MM2H figure, are carried by a citation that is silent on them. One supported row out of ten. The unsupported routes need their own sources (MM2H official portal, Immigration Dept Student Pass rules). ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: FAILS for all four cohorts. No single route permits RESIDE + STUDY + WORK_REMOTE. The near-miss is DE Rantau: it permits RESIDE and WORK_REMOTE for 365 days but is silent on study, and silence is not permission. Cohorts are identical because every route below is nationality-agnostic (MDEC: 'open for application to citizens of all nationalities except Israel'). Per-route sourcing and reasoning: (1) DE RANTAU — quote below is verbatim from the fetched MDEC page, which also lists 'Type of Pass: Professional Visit Pass (Pas Lawatan Ikhtisas (PLIK))' and 'Features: • Stay from 3 up to 12 months • Pass renewable for up to additional 12 months'. maxStayDays=365, NOT 730: the second 12 months is a renewal, and the page words it as an option, not a right, so per the rules it is not counted. It is a work pass; nothing on the page or on https://rai.malaysia.gov.my/digital-services/674d807abf56b738f6c0a469 grants study rights. (2) STUDENT PASS — https://www.imi.gov.my/index.php/en/main-services/pass/student-pass/ grants residence + study, and permits only limited on-site part-time work ('limited to 20 hours per week, only at approved locations' — restaurants, petrol stations, mini markets, hotels). That is not WORK_REMOTE, and the page is silent on remote work for a foreign employer, so it does not permit it. CAVEAT ON THE NUMBER: IMI does not state a fixed Student Pass duration on that page; 365 reflects the one-year-at-a-time renewal cycle described by EMGS, and renewal is conditional (CGPA 2.0, 80% attendance) hence discretionary. Treat 365 as weakly sourced — it does not change the verdict, which turns on permits. (3) MM2H — https://www.imi.gov.my/index.php/en/main-services/malaysia-my-second-home-mmh2-en/ shows a 5-year Social Visit Pass (1825 days) but work is only 'part-time working' subject to separate approval (company authorisation, job justification, local newspaper advertisement) so it is not permitted as of right, and applicants 18 and above must 'apply student pass' separately for study — the IMI page itself confirms MM2H does not carry study rights for adults. RESIDE only. (4) CN VISA-FREE — the China-Malaysia mutual exemption in force since 2025-07-17 is explicitly not a residence/work/study route: the PRC Embassy in Malaysia states 'Residence, employment, study, media activities, or other activities are not covered by the Agreement' (https://my.china-embassy.gov.cn/eng/fwzc/lsyw/qz/202508/t20250801_11681401.htm), 30 days per entry, 90 days per 180. permits=[] is a sourced finding, not a gap. DELIBERATELY OMITTED: the ordinary visa-free social visit for US/EU/IN. IMI's own list confirms these cohorts are not visa-required (see entry-friction), but IMI does not state the permitted duration anywhere I could fetch, and the widely-repeated '90 days' figure appears only on relocation/visa-agency sites. Rather than attach a plausible 90 to a real URL, I left the route out — it is a tourist route and would fail on permits regardless of duration. QUOTE THE RESEARCHER OFFERED: “At its core is the DE Rantau Nomad Pass, a newly introduced Professional Visit Pass that allows qualified foreign digital nomads to stay and work in Malaysia for up to 12 months.” | seed | seed | e0a6b9c5-2c2b-411e-b687-f65e19c749b4 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 70 | eef80e00-8f3a-4a04-a912-a4149f549b31 | my | visa-access | {"route":"DE Rantau Nomad Pass (Professional Visit Pass / Pas Lawatan Ikhtisas)","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":365,"passportSet":"*"} | NULL | https://www.mdec.my/static/pdf/derantau/251105_DE%20Rantau_Pass_FAQ_V8.pdf | Malaysia Digital Economy Corporation (MDEC), Ministry of Digital | 2025-11-05 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from MDEC's DE Rantau Pass FAQ V8 (PDF dated 2025-11-05 by its own filename, 251105): Q1 — “DE Rantau Nomad Pass is a type of Professional Visit Pass newly designed to allow qualified foreign digital nomads to travel and work in Malaysia.” Q4 — “The Professional Visit Pass (Pas Lawatan Ikhtisas) is valid for a period between three (3) to twelve (12) months, with the option to renew for an additional twelve (12) months, allowing a maximum stay of twenty-four (24) months in total.” Q7 — “The pass is open to applicants of all nationalities, with the exception of citizens of Israel.” WHAT IT PERMITS: RESIDE (stay up to 12 months) and WORK_REMOTE (the eligibility test is remote/digital work — remote workers must show an ‘Active employment contract’ with an ‘Employer [that] is foreign/non-Malaysian based companies’, freelancers an active project contract; income floor USD24,000/yr tech, USD60,000/yr non-tech, per the same FAQ and mdec.my/derantau). WHAT IT DOES NOT PERMIT: STUDY — the FAQ is silent on study rights for the principal, and silence is not permission; Q15 confirms even a CHILD dependant “who wish to study in Malaysia, please apply for a Student Pass via the school”, i.e. the pass itself carries no study right. Spouse may not work under the pass (Q10). maxStayDays=365, NOT 730: Q4 words the second 12 months as ‘the option to renew’, which is not an as-of-right grant, so per the per-entry/as-of-right rule it is not counted. PASSPORTSET ‘*’ BASIS (STRONG, WITH ONE CARVE-OUT): this is an AFFIRMATIVE universality statement (‘open to applicants of all nationalities’), not mere absence of a condition — but it carries one published exception, citizens of Israel. None of the scenario's four passport sets (US, EU/EEA/CH, IN, CN) is IL. PASSPORT.md §5's passport-exclusions subtraction from a ‘*’ route is the proper home for the IL carve-out and the gate does not yet read it — recorded here so nobody reads ‘*’ as unconditional. TERRITORIAL LIMIT (Q16): the pass “facilitates your stay only in any states in Peninsular Malaysia and Federal Territory of Labuan” — Forest City (Johor) is in Peninsular Malaysia, so this does not bite the scenario; a separate DE Rantau Sarawak variant exists and was not researched. RENDERING DISCLOSURE (fetch fact): fetch_page on this PDF URL fails with ‘Download is starting’ (NAVIGATION_FAILED); the document was read in full via the Google Docs viewer rendering of this exact URL (11 pages). The live page https://www.mdec.my/derantau and the MyGovernment service page https://rai.malaysia.gov.my/digital-services/674d807abf56b738f6c0a469 (“allows qualified foreign digital nomads to work remotely from Malaysia for up to 12 months, with the option to renew for an additional 12 months”) were both fetched live and corroborate pass type, duration and renewal wording. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Direct fetch_page of the PDF URL fails with NAVIGATION_FAILED ('Download is starting') — same wall the researcher hit. Independently re-read via Google Docs viewer rendering of the exact URL (renders 'Page 1 of 11'; pages 1–3, containing every quoted answer Q1–Q17, rendered in full and match verbatim). Live corroboration fetched independently: https://www.mdec.my/derantau (OK) states 'allows qualified foreign digital nomads to stay and work in Malaysia for up to 12 months' and 'Pass renewable for up to additional 12 months', and names the separate DE Rantau Sarawak variant, all consistent with the claim. Verification points: (1) '*' with the Israel carve-out is exactly what Q7 states — an affirmative universality statement with one named exception, and the claim's note discloses it; (2) permits RESIDE+WORK_REMOTE match — the pass's own purpose is remote/digital work (eligibility: employer 'is foreign/non-Malaysian based companies', freelance project contracts), and no study right is stated anywhere on pages 1–3 (Q15 confirms even a child dependant needs a separate Student Pass); (3) maxStayDays=365 is correct under the as-of-right rule — Q4's second 12 months is 'the option to renew', not an automatic grant, so 730 would have been the defect and 365 is the honest figure; the pass is multi-entry (Q12) and is itself the stay authorisation, so 365 is not a visa-validity conflation. | seed | seed | 56109e6b-67c1-40bb-81ad-88c049ecd380 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 71 | 0c4518d4-7b2a-43eb-8a2c-4783b8d23b0a | my | visa-access | {"route":"Student Pass","permits":["RESIDE","STUDY"],"maxStayDays":365,"passportSet":"*"} | NULL | https://web.archive.org/web/20260220063822/https://www.imi.gov.my/index.php/en/main-services/pass/student-pass/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-02-20 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the page as captured 2026-02-20: “All foreigners who wish to pursue their education in any level of education in Malaysia are required to apply for a Student Pass from the Department of Immigration Malaysia.” — “Student Pass – RM60.00 per year or part thereof” — “Dependants of the Student Pass will be issued a Dependant Pass valid for 12 months according to the validity of the Student Pass of the principal.” WHAT IT PERMITS: RESIDE + STUDY. The first quote also establishes the converse that matters for this gate: study in Malaysia REQUIRES a Student Pass, so no other route can carry a study right for a foreign adult. WHAT IT DOES NOT PERMIT: WORK_REMOTE — the page as captured states no work right of any kind; silence is not permission. (The 20-hours-per-week on-site part-time rule cited by the previous pass does NOT appear on this capture and is not claimed here.) maxStayDays=365 IS WEAKLY SOURCED AND FLAGGED: IMI states no fixed principal-pass duration; 365 is inferred from the per-year fee line and the dependant-pass sentence tying a 12-month Dependant Pass to ‘the validity of the Student Pass of the principal’, i.e. a one-year-at-a-time cycle. Renewal is conditional (EMGS publishes CGPA/attendance conditions) and therefore discretionary — NOT counted beyond one year. I could not fetch an official page stating the renewal conditions this pass: educationmalaysia.gov.my's renewal news page is a JS shell that rendered empty, and visa.educationmalaysia.gov.my/guidelines & /guidelines/required-documents render only accordion headers (fetch facts, recorded in notes). PASSPORTSET ‘*’ BASIS (WEAKER FORM — stated scope, not an affirmative neutrality statement): the page's own scope is ‘All foreigners’; visa-required nationals additionally need a Visa With Reference (VDR) before entry, which is procedure, not a nationality bar. I did not find an affirmative nationality-neutrality statement, and Malaysia's general posture toward Israeli nationals (Ministry of Home Affairs approval to enter) is not addressed on this page — both said explicitly so ‘*’ is not over-read. TIER DISCLOSURE (fetch fact): imi.gov.my refuses this client IP wholesale (WAF ‘Web Page Blocked! … Attack ID: 20000051’ on every path tried, including the site root) — that is a refusal of THIS fetcher, not evidence about the source. The URL actually read is the Internet Archive's rendering of IMI's own page, tiered as the mirror it is rather than as OFFICIAL_PRIMARY; content authorship remains the Immigration Department. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Internet Archive capture (2026-02-20) re-fetched OK; all three quotes verbatim on the page. Live https://www.imi.gov.my/index.php/en/main-services/pass/student-pass/ attempted independently: WAF-blocked to this fetcher too (HTTP 500, 'Web Page Blocked!', Attack ID 20000051) — the mirror basis is a fetch fact shared by both parties, not a defect; ACADEMIC_OR_INDEX tiering of the mirror is correct. Verification points: (1) permits RESIDE+STUDY match; no work right of any kind appears on the captured page, so WORK_REMOTE's absence is correct (the claim's own disclosure that the 20-hour part-time rule does NOT appear on this capture is accurate — I note a 'PROCEDURE FOR APPLYING FOR WORK PERMISSION ENDORSEMENT' menu entry exists elsewhere on IMI's site, an on-site-work endorsement surface, not a remote-work right, and not on this page); (2) maxStayDays=365 is an inference, exactly as the claim flags — the page states no fixed principal-pass duration; the per-year fee line and the dependant-pass sentence tying a 12-month Dependant Pass to 'the validity of the Student Pass of the principal' jointly support a one-year-at-a-time cycle, and no stronger figure exists on the page; the claim's self-flagging ('WEAKLY SOURCED AND FLAGGED') is the honest recording and renewal is correctly not counted; (3) '*' as stated scope ('All foreigners') rather than affirmative neutrality is exactly what the page supports, and the claim says so. | seed | seed | ffde7f67-8eb0-4adb-a919-b5f844263cbe | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 72 | 8e4a4879-ca99-4a30-ac0b-f7929989a2d4 | my | visa-access | {"route":"Malaysia My Second Home (MM2H) — Silver category","permits":["RESIDE"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://www.mm2h.gov.my/category/silver | Ministry of Tourism, Arts and Culture (MOTAC), official MM2H portal | 2026-02-10 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the live MOTAC page on the retrieval date: “MM2H pass for a period of 5 years with Multiple Entry Visa (MEV)” — “BUSINESS/INVESTMENT ACTIVITIES Not allowed. CAREER OPPORTUNITIES Not allowed.” — “EDUCATION Dependent/s (children) is allowed to pursue their education up to tertiary level… allowed to use the existing MM2H or be given a Student Pass automatically.” PERMITS: RESIDE only. STUDY is granted to child dependants, not to the principal — a scenario member (adult principal) gets no study right. Work of any kind is expressly ‘Not allowed’. Conditions: USD150,000 fixed deposit, compulsory purchase of an RM600,000+ residence, minimum presence 90 days (cumulative) per year. maxStayDays=1825 records the 5-year pass period (5×365): the MM2H pass is itself the stay authorisation (MOTAC's portal describes it as ‘a renewable social visit pass with multiple entry-visa’), so pass period, per-entry stay and validity coincide here — this is NOT a visa-validity-for-stay conflation; renewal (‘can be renewed for every 5 years’) requires fresh documents and is not counted. PASSPORTSET ‘*’ BASIS (AFFIRMATIVE STATEMENT WITH A PUBLISHED CONDITION): the guidelines page (https://www.mm2h.gov.my/apply/guidelines, fetched live) states “Participation into the MM2H Program is open to every eligible foreign individual from sovereign countries that has diplomatic relations with Malaysia.” That is affirmative openness subject to one nationality-shaped condition (diplomatic relations — which excludes e.g. Israel, with which Malaysia has none). That all four scenario passport sets are states holding diplomatic relations with Malaysia is an analyst premise, not stated on the page — recorded so ‘*’ is not over-read; PASSPORT.md §5 passport-exclusions is the proper home for the subtraction. FETCH FACT: mm2h.gov.my now resolves fully to this fetcher (the 2026-07-17 pass's 403 was the old fetcher, not the site); the /apply/guidelines page renders only accordion section headers (sections 01–14), so category detail is sourced from the category pages, which render in full. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live https://www.mm2h.gov.my/category/silver fetched OK (Last Update: 10/02/2026, matching effectiveAt); every quote verbatim. https://www.mm2h.gov.my/apply/guidelines also fetched OK — the general-requirements openness sentence renders verbatim in the page body (not only accordion headers), so the '*' basis quote is independently verified, diplomatic-relations condition and all. Verification points: (1) permits RESIDE only is correct — work expressly 'Not allowed', education row grants children only, so no STUDY for a principal; (2) maxStayDays=1825 = the 5-year pass period; the page states no separate per-entry cap distinct from pass validity and MOTAC treats the pass itself as the stay authorisation, so this is not a visa-validity conflation; renewal ('renewed for every 5 years… latest documents') correctly not counted; (3) USD150,000 FD / RM600,000 residence / 90-day presence conditions all on the page as recorded. | seed | seed | 8a06a09f-269e-4b00-9cf5-acc63ae8b3d5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 73 | 3e620779-19b3-4f34-beae-f2bf2b7e1648 | my | visa-access | {"route":"Malaysia My Second Home (MM2H) — Gold category","permits":["RESIDE"],"maxStayDays":5475,"passportSet":"*"} | NULL | https://www.mm2h.gov.my/category/gold | Ministry of Tourism, Arts and Culture (MOTAC), official MM2H portal | 2026-02-10 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the live MOTAC page on the retrieval date: “MM2H pass for a period of 15 years with Multiple Entry Visa (MEV)” — “BUSINESS/INVESTMENT ACTIVITIES Not allowed. CAREER OPPORTUNITIES Not allowed.” PERMITS: RESIDE only; education row grants child dependants only, so no STUDY for the principal; work expressly not allowed. Conditions: USD500,000 fixed deposit, compulsory RM1 million+ residence purchase, minimum 90 days presence per year. maxStayDays=5475 records the 15-year pass period (15×365) — the MM2H pass is itself the stay authorisation (see Silver claim for the reasoning; same portal, same wording pattern). PASSPORTSET ‘*’ BASIS: identical to the Silver claim — the shared guidelines page states “Participation into the MM2H Program is open to every eligible foreign individual from sovereign countries that has diplomatic relations with Malaysia”; affirmative, with the diplomatic-relations condition disclosed; scenario sets unaffected (analyst premise). effectiveAt is the portal's own ‘Last Update: 10/02/2026’ (read DD/MM/YYYY); the underlying policy dates from the 2024 MM2H revamp whose gazetted instrument I did not locate this pass — disclosed rather than guessed. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live https://www.mm2h.gov.my/category/gold fetched OK; quotes verbatim; footer 'Last Update: 10/02/2026' matches effectiveAt. permits RESIDE only correct (work expressly not allowed; study for child dependants only). maxStayDays=5475 = 15-year pass period, same pass-is-the-stay-authorisation reasoning as Silver, no separate per-entry cap stated on the page. USD500,000 FD and RM1 million residence conditions on the page as recorded. '*' basis verified on the live guidelines page (see claim 2). | seed | seed | 7a223d48-7d61-468b-b2fc-2df2e8de2869 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 74 | 8bf8c767-bcd5-47ae-b7ad-b7d0681cf567 | my | visa-access | {"route":"Malaysia My Second Home (MM2H) — Platinum category","permits":["RESIDE"],"maxStayDays":7300,"passportSet":"*"} | NULL | https://www.mm2h.gov.my/category/platinum | Ministry of Tourism, Arts and Culture (MOTAC), official MM2H portal | 2026-02-10 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the live MOTAC page on the retrieval date: “20 years MM2H pass for a period of 20 years with Multiple Entry Visa (MEV)” — “BUSINESS/INVESTMENT ACTIVITIES Permissible. CAREER OPPORTUNITIES Permissible.” — “EDUCATION Dependent/s (children) is allowed to pursue their education up to tertiary level…” PERMITS: RESIDE. THE ONE AMBIGUITY, RECORDED RATHER THAN RESOLVED IN OUR FAVOUR: Platinum — alone among the four categories — marks business/investment activities and ‘career opportunities’ as ‘Permissible’, but the page states neither the scope nor the mechanism (as-of-right vs. separate approval), and says nothing about remote work for a foreign employer. The guidelines accordion section ‘10. Business/Investment Activities/Career Opportunities’ did not render to the fetcher, so the detail could not be read. WORK_REMOTE is therefore NOT recorded — permits names only what the source grants, and a one-word ‘Permissible’ with undefined scope is not a grant of the scenario's activity. STUDY: child dependants only; not the principal. Conditions: USD1 million fixed deposit, RM200,000 participation fee, compulsory RM2 million+ residence, 90 days/year minimum presence. maxStayDays=7300 records the 20-year pass period (20×365); the MM2H pass is itself the stay authorisation (see Silver claim). PASSPORTSET ‘*’ BASIS: identical to Silver — guidelines quote “open to every eligible foreign individual from sovereign countries that has diplomatic relations with Malaysia”, affirmative with the condition disclosed. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live https://www.mm2h.gov.my/category/platinum fetched OK; quotes verbatim, including the odd doubled '20 years… for a period of 20 years' wording, which the claim reproduced faithfully. The decisive judgment call verified: the page marks business/investment and 'career opportunities' as one-word 'Permissible' with no stated scope or mechanism, and says nothing about remote work for a foreign employer — I independently confirmed the guidelines accordion section '10. Business/Investment Activities/Career Opportunities' renders as a bare header with no body text to this fetcher too, so the detail is unreadable on the portal. Withholding WORK_REMOTE is therefore correct, and this is NOT an under-claim to correct upward: 'Permissible' expressly grants some work-shaped capacity but not the scenario's activity, and permits may only name what the source grants. If the permits vocabulary ever carries a local-work/business activity, the gate owner has an express one-word basis to weigh — recorded here so the upward possibility is visible. maxStayDays=7300 = 20-year pass period, same basis as Silver. USD1 million FD, RM200,000 fee, RM2 million residence, 90 days/year all on the page as recorded. | seed | seed | ac4baf81-4fd9-4cd3-aa7d-9cef6b7ff54e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 75 | ab9d9a44-6540-4ddf-ab85-b57f62134879 | my | visa-access | {"route":"Malaysia My Second Home (MM2H) — SEZ/SFZ category (Forest City, Johor)","permits":["RESIDE"],"maxStayDays":3650,"passportSet":"*"} | NULL | https://www.mm2h.gov.my/category/sez | Ministry of Tourism, Arts and Culture (MOTAC), official MM2H portal | 2026-02-10 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the live MOTAC page on the retrieval date: “MM2H pass for a period of 10 years with Multiple Entry Visa (MEV)” — “Compulsory to purchase and own a property/house* in Forest City, Johor after obtaining the approval as MM2H participant.” — “BUSINESS/INVESTMENT ACTIVITIES Not allowed. CAREER OPPORTUNITIES Not allowed.” PERMITS: RESIDE only; education row grants child dependants only; work expressly not allowed — note the irony that the category built FOR Forest City (this scenario's calibration site) prohibits work outright, so it cannot carry a WORK_REMOTE member. Conditions: fixed deposit USD65,000 (ages 21–49) or USD32,000 (50+), RM1,000 fee, compulsory Forest City property purchase (floor price subject to Johor state policy), 90 days/year minimum presence. maxStayDays=3650 records the 10-year pass period (10×365); the MM2H pass is itself the stay authorisation (see Silver claim). PASSPORTSET ‘*’ BASIS: identical to Silver — guidelines quote “open to every eligible foreign individual from sovereign countries that has diplomatic relations with Malaysia” (SEZ/SFZ minimum age 21 rather than 25). FETCH FACT: the URL is /category/sez — /category/sez-sfz 404s. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live https://www.mm2h.gov.my/category/sez fetched OK (title: 'Category Special Economic Zone / Special Financial Zone'); quotes verbatim. permits RESIDE only correct — the Forest City category expressly prohibits work, exactly as the claim notes. maxStayDays=3650 = 10-year pass period, same basis as Silver. FD USD65,000 (21–49) / USD32,000 (50+), RM1,000 fee, Johor floor-price footnote, 90 days/year — all on the page as recorded. Guidelines page confirms SEZ/SFZ minimum age 21 ('aged 21 years old and above for the SEZ/SFZ category'). | seed | seed | f9833a90-daf0-466c-839d-d731bd4c278f | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 76 | 895e45f9-2b7a-40c6-9295-92ee02ac5934 | my | visa-access | {"route":"China–Malaysia Mutual Visa Exemption Agreement","permits":["VISIT"],"maxStayDays":30,"passportSet":"CN"} | NULL | https://my.china-embassy.gov.cn/eng/fwzc/lsyw/qz/202508/t20250801_11681401.htm | Embassy of the People's Republic of China in Malaysia | 2025-07-17 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the embassy's FAQ on the Agreement (page dated 2025-08-01; Agreement “has come into force on July 17, 2025”): “holders of Chinese passports for public affairs and ordinary passports… may enter into, exit from, or transit through the territory of the other Contracting Party without a visa and stay for a period of stay not exceeding 30 days… for the purposes of vacation/tour, family and friends visit, business, exchange, private affairs, medical treatment and international traffic (crew members). The cumulative duration of stay must not exceed 90 days within each period of 180 days.” And decisively for this gate: “Residence, employment, study, media activities, or other activities are not covered by the Agreement.” PERMITS: VISIT only — the listed purposes (tourism, family visit, business, exchange, private affairs, medical) are all visit-shaped, and the exclusion of residence/employment/study is a SOURCED finding, not a gap. maxStayDays=30 per entry, with a 90-in-180 cumulative cap and a 5-year term auto-renewing for 5 more (Q8). TIER: the PRC embassy is a state party to the bilateral agreement restating its terms — OFFICIAL_SECONDARY relative to the instrument itself, which I did not locate in gazette form. The Malaysian side corroborates the exemption's existence: CN is absent from IMI's visa-required list (see the US/IN claims' source) and IMI's site carries a ‘FAQ Malaysia – China Mutual Visa Exemption’ page I could not read (imi.gov.my WAF-blocks this client). This route fails every scenario activity and the 180-day floor; it is recorded so the CN cohort's tourist access is never mistaken for admission (BRIEF §6). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Live embassy page fetched OK (dated 2025-08-01); every quote verbatim, including the residence/employment/study exclusion sentence checked word-for-word. permits [VISIT] matches — all listed purposes are visit-shaped and the exclusion of the three scenario activities is the page's own affirmative statement, not an inference. maxStayDays=30 matches Q4's 'up to 30 days per entry' (expressly per-entry, so no validity conflation; the 90-in-180 cumulative cap is disclosed in the note, not smuggled into the stay figure). effectiveAt 2025-07-17 matches. OFFICIAL_SECONDARY tiering is right — a state party restating the bilateral instrument. Malaysian-side corroboration independently verified: China is absent from IMI's 30-country visa-required list on the 2026-07-07 capture, and IMI's site carries a 'FAQ Malaysia – China Mutual Visa Exemption' menu entry. | seed | seed | 7f66ea4f-b155-44cb-a584-f17b78cd9a3f | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 77 | 8b516c52-f72b-4e53-a8c0-040f45504100 | my | visa-access | {"route":"Visa exemption for Indian nationals (expires 31 December 2026)","permits":[],"maxStayDays":null,"passportSet":"IN"} | NULL | https://web.archive.org/web/20260707135011/https://www.imi.gov.my/index.php/en/main-services/visa/visa-requirement-by-country/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-07-07 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE, verbatim from IMI's Visa Requirement by Country page as captured 2026-07-07: India appears on the list of “Countries required to apply for a visa to enter Malaysia” as “India**”, and the page's own NOTE overrides it: “India** citizen: visa exempts until 31st December 2026.” WHAT THIS ESTABLISHES: an IN passport currently needs no visa to enter Malaysia — and nothing more. permits=[] because the page grants no activity: it states neither the purpose scope nor the permitted duration of the visa-exempt entry. maxStayDays=null DELIBERATELY: the permitted stay appears on no official page I could fetch; the ubiquitous ‘30/90 days’ figures trace only to relocation and visa-agency sites (excluded tier), so the number is omitted rather than borrowed — the route fails the gate on permits regardless. HARD DATE: the exemption expires 2026-12-31, ~5 months out, and IMI states no successor; absent it, India reverts to the visa-required list. Note also that India is NOT on the page's separate 51-country table of countries that “do not require a visa for any purpose of entry”, so for study/work purposes an Indian national requires a Visa With Reference (VDR/eVAL) — same page, section 2. TIER DISCLOSURE (fetch fact): the live imi.gov.my refuses this client IP (WAF, Attack ID 20000051, all paths including the root); the URL read is the Internet Archive rendering of IMI's own page, tiered as the mirror; content authorship is IMI. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Internet Archive capture (2026-07-07) re-fetched OK; quote verbatim. Live imi.gov.my WAF-blocked to this fetcher too (Attack ID 20000051) — mirror basis stands as a shared fetch fact. The two deliberate absences verified as real: the page states NO purpose scope and NO permitted duration for this exemption, so permits=[] and maxStayDays=null are the correct floor — any 30/90-day figure would have been borrowed from an excluded tier, exactly as the claim says. Also verified: India is NOT on the page's 51-country 'do not require a visa for any purpose of entry' table (checked all 51 rows), so the VDR-for-study/work point holds; the 2026-12-31 expiry is the page's own hard date with no successor stated. | seed | seed | fcbb8211-2574-4eea-bb53-b638b20d8af8 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 78 | d0d4dd17-944c-4c6b-bbfa-6c42cf9800de | my | visa-access | {"route":"Visa-free entry for social visits (absent from IMI's visa-required list)","permits":[],"maxStayDays":null,"passportSet":"US"} | NULL | https://web.archive.org/web/20260707135011/https://www.imi.gov.my/index.php/en/main-services/visa/visa-requirement-by-country/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-07-07 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | BASIS — ABSENCE FROM A COMPLETE PUBLISHED LIST, READ IN FULL: IMI's page enumerates the “Countries required to apply for a visa to enter Malaysia (List of involved countries)” — 30 entries, Afghanistan through Sri Lanka, quoted in full in researcher-notes.md — and the United States is not on it, therefore a US passport is not visa-required for entry. WHAT THIS DOES NOT ESTABLISH: purpose scope or duration. permits=[] because the page grants no activity; maxStayDays=null DELIBERATELY — the permitted visa-free stay for US nationals appears on no official Malaysian page I could fetch, and the widely repeated ’90 days’ exists only on relocation/visa-agency sites (excluded tier), so it is omitted rather than borrowed; the route fails the gate on permits regardless. THE NUANCE THAT MATTERS FOR THIS GATE (same page, section 2): the US is also NOT on the separate 51-country table of countries that “do not require a visa for any purpose of entry” (Switzerland, Netherlands, Ireland, Cyprus, Malta, Liechtenstein and the UK are; most EU states and the US are not) — so for the purpose of “studying, working, or other purposes” a US national requires a Visa With Reference (VDR/eVDR) or Visa Approval Letter subject to approval. Tourist visa-freedom is affirmatively NOT activity access here, in the authority's own structure — BRIEF §6 made literal. TIER DISCLOSURE (fetch fact): live imi.gov.my WAF-blocks this client IP (Attack ID 20000051, every path tried); the URL read is the Internet Archive rendering of IMI's own page; content authorship is IMI. The 2026-07-17 pass fetched the live page by curl and its verbatim list matches this capture exactly — independent corroboration of the rendering's fidelity. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Internet Archive capture (2026-07-07) re-fetched OK. I independently counted the visa-required list: 30 entries, US absent — the absence-from-a-complete-published-list basis is real and the list was re-read in full. I also re-read all 51 rows of the any-purpose table: US absent, so a US national needs a Visa With Reference (VDR/eVDR) or Visa Approval Letter for 'studying, working, or other purposes' — the page's own structure, as claimed. permits=[] and maxStayDays=null verified as genuine absences: the page states no duration and no purpose scope for US visa-free entry, and no official figure exists on this page to record. Live imi.gov.my WAF-blocked to this fetcher too (Attack ID 20000051); mirror basis stands. | seed | seed | a5c2c9f6-0722-4011-9b46-f2b52034c871 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 79 | db741539-c449-47f9-a531-9ebef5192574 | my | visa-access | {"passportSet":"US","routesEnumerated":true} | NULL | https://web.archive.org/web/20260210001748/https://www.imi.gov.my/index.php/en/main-services/pass/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-02-10 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | ENUMERATION BASIS — FOUR DOCUMENTS JOINTLY CLOSE THE LIST (per the multi-document enumeration rule): (1) THIS PAGE — IMI's Main Services » Pass index, the authority's own complete published list of pass types, verbatim: “MACS | Visitor's Pass (Temporary Employment) | Employment Pass | Social Visitor Pass | Long Term Social Visit Pass | Short Term Social Visit Pass | Professional Visitor Pass | Student Pass | Residence Pass”, with ‘Entry Permit’, ‘Foreign Worker’, ‘Foreign Domestic Helper (FDH)’ and ‘Malaysia My Second Home (MMH2)’ listed alongside under Main Services; (2) IMI's Visa Requirement by Country page (capture 2026-07-07) for the entry/visa-exemption routes; (3) MOTAC's MM2H portal (live) for the four MM2H categories; (4) MDEC's DE Rantau FAQ V8 for the Professional Visit Pass nomad variant. WHY NO ENUMERATED ROUTE CAN PERMIT ALL THREE SCENARIO ACTIVITIES: IMI's Student Pass page states “All foreigners who wish to pursue their education in any level of education in Malaysia are required to apply for a Student Pass” — so STUDY is available only via the Student Pass, and the Student Pass carries no remote-work right; therefore no single route on the authority's list grants RESIDE+STUDY+WORK_REMOTE. DISCLOSURES, IN FULL: (a) pass pages NOT individually fetched this pass: MACS (a border-clearance facility, not a stay route), Visitor's Pass (Temporary Employment) (sector-quota labour from approved source countries), Social Visitor Pass, Long/Short Term Social Visit Pass (family-tie and visit routes), Residence Pass, Foreign Worker, FDH — assessed from the list and each route's published scope only; the Employment Pass page WAS fetched (capture 2026-03-07: “issuance of passes to Expatriates who have been approved by the Expatriate Committee” — local employment, no study right); (b) ENTRY PERMIT (permanent residence) is on the list and was NOT researched: it is discretionary, sponsor-gated, and not a published as-of-right route to a 180-day stay for a newly arriving member, but a PR would in principle permit all activities — the gate owner must weigh whether this weakens the enumeration before licensing a FAIL; (c) a DE Rantau Sarawak regional variant exists (MDEC) and was not separately researched; (d) the route list is nationality-neutral — this enumeration is recorded once per scenario passport set because the gate matches per-set strings (gates.ts), and all four enumeration claims rest on these SAME cited documents; the D-036 one-source-one-claim tension is acknowledged and disclosed rather than hidden. TIER DISCLOSURE (fetch fact): live imi.gov.my WAF-blocks this client IP; URL read is the Internet Archive rendering. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: All four joint-basis documents independently re-fetched: (1) the Pass index capture — the Passes list matches the claim's verbatim quote exactly; (2) the Visa Requirement by Country capture (OK, matches); (3) MOTAC's MM2H portal live (OK, all four categories match); (4) MDEC's DE Rantau FAQ V8 via Google Docs viewer (matches). VERIFIER'S JUDGMENT ON CLOSURE, stated explicitly as asked: the enumeration holds AS A LIST — the Pass index is the authority's own published inventory of pass types and my re-fetch reproduces it exactly, so routesEnumerated:true is backed by the authority's actual complete list. The Entry Permit admission does NOT defeat the enumeration: the Entry Permit is ON the list, named, and disclosed as unresearched, so the list remains complete even though one listed route's interior is unread. What the admission DOES weaken is the downstream closure argument ('no single route permits RESIDE+STUDY+WORK_REMOTE') — that argument is proven for every route whose scope is sourced, but for the Entry Permit it rests on the researcher's own disclosed characterisation, and a PR would in principle permit all activities; the claim says exactly this and hands the weighing to the gate owner, which is the correct recording. TWO SURFACES THE CLAIM DID NOT NAME, found on my re-fetch of the same capture and recorded so the verifier's read is on the record: a 'FAQ – Graduate (Social) Visit Pass' menu entry and a 'PROCEDURE FOR APPLYING FOR WORK PERMISSION ENDORSEMENT' menu entry. Neither appears in the Passes index itself; the Graduate pass is by its own title a (Social) Visit Pass variant within a listed family, and the work-permission endorsement is an on-site-work procedure, not a route — so neither adds an unlisted route class, but a future pass should read both before this enumeration licenses a FAIL. HELD AT NEEDS_EVIDENCE BY THE MERGE, 2026-07-18 — deliberately NOT pass-eligible, so this enumeration cannot license a FAIL yet. Both the researcher and the adversarial verifier flagged the same gap: the enumeration holds as a LIST, but the Entry Permit (PR) — a listed, in-principle qualifier — was never researched for its permits, and the verifier surfaced two unread menu entries ('FAQ – Graduate (Social) Visit Pass' and a work-permission-endorsement procedure). D-036's rule is that you cannot fail a cohort without evidence you read ALL of it, and 'all of it' includes each listed route's permits: a bar computed while a listed route's permits are unknown converts disclosed-not-proven into FAIL. UPGRADE PATH: read the Entry Permit conditions and the two menu entries; if none yields a qualifying route, raise this to CORROBORATED and the gate's FAIL becomes licensed. UPGRADED TO CORROBORATED BY THE MERGE, 2026-07-18 — the holdback is discharged. The three listed routes that were unread when this was held (Entry Permit/PR, Graduate (Social) Visit Pass, LTSVP-spouse + work-permission endorsement) have now been read from primary sources and adversarially verified (2 CONFIRMED, 1 CORRECTED, 0 REFUTED; evidence/visa-my/followup-enum-*). None permits RESIDE+STUDY+WORK_REMOTE in one route: Entry Permit lacks STUDY (Immigration Act 1959/63 silent on study for PR holders; the verifier independently confirmed no primary source grants or denies it, so silence is read conservatively as no grant), the Graduate pass lacks WORK_REMOTE, the spouse LTSVP lacks STUDY. THE COMPLETENESS DETERMINATION, made explicitly: the FAIL rests on a STRUCTURAL bar, not merely a compiled list — STUDY in Malaysia runs only through the Student Pass, which carries no remote-work right, so no single instrument can combine study with remote work regardless of list-completeness. RESIDUAL DISCLOSED: the Student-Pass-monopoly sentence the researcher quoted is no longer verbatim on the live IMI page (it now reads 'granted to non-citizen students ... who pursue studies'); the verifier judged the substance unchanged. This licenses Malaysia's visa-access FAIL, which excludes the Forest City candidate — the consequence the holdback existed to gate. Reversal path if the structural bar is later shown wrong (an unlisted route granting STUDY+WORK_REMOTE): drop these enumerations back to NEEDS_EVIDENCE. | seed | seed | e1c8585d-98cc-41fc-bb95-d3b4ee03086f | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 80 | 1ab93d14-2cbd-43f1-99fc-38ec9458a91c | my | visa-access | {"passportSet":"EU/EEA/CH","routesEnumerated":true} | NULL | https://web.archive.org/web/20260210001748/https://www.imi.gov.my/index.php/en/main-services/pass/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-02-10 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Identical basis, closure argument, and disclosures to the US enumeration claim — four documents jointly close the authority's route list (IMI Pass index quoted verbatim there; IMI Visa Requirement by Country; MOTAC MM2H portal; MDEC DE Rantau FAQ V8), and IMI's Student Pass page makes the Student Pass the only study route (“All foreigners who wish to pursue their education in any level of education in Malaysia are required to apply for a Student Pass”), which carries no remote-work right. ONE EU/EEA/CH-SPECIFIC ADDITION: Malaysia does not publish a single rule for an ‘EU/EEA/CH’ bloc. On IMI's visa page the bloc's members are scattered — no EU state is visa-required (Montenegro and Serbia are on the list but are not EU members), while Switzerland, Netherlands, Ireland, Cyprus, Malta and Liechtenstein additionally appear on the 51-country ‘no visa for any purpose of entry’ table and the rest do not. I therefore emitted NO tourist-entry route claim for this passport set: keying one ‘EU/EEA/CH’ claim would coin a bloc the authority does not use for that rule (PASSPORT.md §1), and thirty per-state claims would multiply one source. The set's coverage comes from the ‘*’ routes (DE Rantau, Student Pass, MM2H ×4), which compose into it. The route LIST, however, is nationality-neutral, so the enumeration holds for this set on the same basis as the others. Same tier disclosure: live imi.gov.my WAF-blocks this client IP; URL read is the Internet Archive rendering. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Same four-document basis re-fetched and verified (see claimIndex 9, including the closure judgment and the two unnamed menu surfaces). The EU/EEA/CH-specific scatter independently verified against the 2026-07-07 capture: both statements are exactly right, and the researcher's decision to emit NO tourist-entry claim keyed 'EU/EEA/CH' is the correct application of PASSPORT.md §1 — Malaysia publishes no rule for that bloc, so a bloc-keyed entry claim would coin an abstraction the authority does not use. The route LIST is nationality-neutral on the source, so the enumeration holds for this set on the same basis. HELD AT NEEDS_EVIDENCE BY THE MERGE, 2026-07-18 — deliberately NOT pass-eligible, so this enumeration cannot license a FAIL yet. Both the researcher and the adversarial verifier flagged the same gap: the enumeration holds as a LIST, but the Entry Permit (PR) — a listed, in-principle qualifier — was never researched for its permits, and the verifier surfaced two unread menu entries ('FAQ – Graduate (Social) Visit Pass' and a work-permission-endorsement procedure). D-036's rule is that you cannot fail a cohort without evidence you read ALL of it, and 'all of it' includes each listed route's permits: a bar computed while a listed route's permits are unknown converts disclosed-not-proven into FAIL. UPGRADE PATH: read the Entry Permit conditions and the two menu entries; if none yields a qualifying route, raise this to CORROBORATED and the gate's FAIL becomes licensed. UPGRADED TO CORROBORATED BY THE MERGE, 2026-07-18 — the holdback is discharged. The three listed routes that were unread when this was held (Entry Permit/PR, Graduate (Social) Visit Pass, LTSVP-spouse + work-permission endorsement) have now been read from primary sources and adversarially verified (2 CONFIRMED, 1 CORRECTED, 0 REFUTED; evidence/visa-my/followup-enum-*). None permits RESIDE+STUDY+WORK_REMOTE in one route: Entry Permit lacks STUDY (Immigration Act 1959/63 silent on study for PR holders; the verifier independently confirmed no primary source grants or denies it, so silence is read conservatively as no grant), the Graduate pass lacks WORK_REMOTE, the spouse LTSVP lacks STUDY. THE COMPLETENESS DETERMINATION, made explicitly: the FAIL rests on a STRUCTURAL bar, not merely a compiled list — STUDY in Malaysia runs only through the Student Pass, which carries no remote-work right, so no single instrument can combine study with remote work regardless of list-completeness. RESIDUAL DISCLOSED: the Student-Pass-monopoly sentence the researcher quoted is no longer verbatim on the live IMI page (it now reads 'granted to non-citizen students ... who pursue studies'); the verifier judged the substance unchanged. This licenses Malaysia's visa-access FAIL, which excludes the Forest City candidate — the consequence the holdback existed to gate. Reversal path if the structural bar is later shown wrong (an unlisted route granting STUDY+WORK_REMOTE): drop these enumerations back to NEEDS_EVIDENCE. | seed | seed | 7c9901dc-60a0-4b55-a5ea-dfc9f89f5084 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 81 | aedc1bfb-919a-4dcd-9fba-73f141db2810 | my | visa-access | {"passportSet":"IN","routesEnumerated":true} | NULL | https://web.archive.org/web/20260210001748/https://www.imi.gov.my/index.php/en/main-services/pass/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-02-10 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Identical basis, closure argument, and disclosures to the US enumeration claim — four documents jointly close the authority's route list (IMI Pass index quoted verbatim there; IMI Visa Requirement by Country; MOTAC MM2H portal; MDEC DE Rantau FAQ V8), and IMI's Student Pass page makes the Student Pass the only study route, which carries no remote-work right. IN-SPECIFIC ADDITIONS: (a) the current visa exemption for Indian nationals expires 2026-12-31 with no stated successor (see the IN route claim) — after that date India reverts to the visa-required list, which changes entry friction but adds no route and removes none, so the enumeration is unaffected; (b) IMI's menu carries an ‘eNTRI for Indian Nationals’ entry (an electronic travel registration facility) which I could not read (imi.gov.my WAF-blocks this client) — eNTRI is an entry-facilitation channel for short social visits, not a residence/study/work route, but it is named here so the record shows it was seen and not silently dropped. Same tier disclosure: live imi.gov.my WAF-blocks this client IP; URL read is the Internet Archive rendering. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Same four-document basis re-fetched and verified (see claimIndex 9, including the closure judgment and the two unnamed menu surfaces). The 'eNTRI for Indian Nationals' menu entry is present on my re-fetch of the capture — the claim's disclosure that it was seen but unreadable (live site WAF-blocked, confirmed blocked to this fetcher too) is accurate; eNTRI is an entry-facilitation surface, not a residence/study/work route, so it does not disturb the enumeration. The IN-specific point that the 2026-12-31 exemption expiry adds no route and removes none is sound: the exemption is an entry-friction fact, and the pass list is nationality-neutral on the source. HELD AT NEEDS_EVIDENCE BY THE MERGE, 2026-07-18 — deliberately NOT pass-eligible, so this enumeration cannot license a FAIL yet. Both the researcher and the adversarial verifier flagged the same gap: the enumeration holds as a LIST, but the Entry Permit (PR) — a listed, in-principle qualifier — was never researched for its permits, and the verifier surfaced two unread menu entries ('FAQ – Graduate (Social) Visit Pass' and a work-permission-endorsement procedure). D-036's rule is that you cannot fail a cohort without evidence you read ALL of it, and 'all of it' includes each listed route's permits: a bar computed while a listed route's permits are unknown converts disclosed-not-proven into FAIL. UPGRADE PATH: read the Entry Permit conditions and the two menu entries; if none yields a qualifying route, raise this to CORROBORATED and the gate's FAIL becomes licensed. UPGRADED TO CORROBORATED BY THE MERGE, 2026-07-18 — the holdback is discharged. The three listed routes that were unread when this was held (Entry Permit/PR, Graduate (Social) Visit Pass, LTSVP-spouse + work-permission endorsement) have now been read from primary sources and adversarially verified (2 CONFIRMED, 1 CORRECTED, 0 REFUTED; evidence/visa-my/followup-enum-*). None permits RESIDE+STUDY+WORK_REMOTE in one route: Entry Permit lacks STUDY (Immigration Act 1959/63 silent on study for PR holders; the verifier independently confirmed no primary source grants or denies it, so silence is read conservatively as no grant), the Graduate pass lacks WORK_REMOTE, the spouse LTSVP lacks STUDY. THE COMPLETENESS DETERMINATION, made explicitly: the FAIL rests on a STRUCTURAL bar, not merely a compiled list — STUDY in Malaysia runs only through the Student Pass, which carries no remote-work right, so no single instrument can combine study with remote work regardless of list-completeness. RESIDUAL DISCLOSED: the Student-Pass-monopoly sentence the researcher quoted is no longer verbatim on the live IMI page (it now reads 'granted to non-citizen students ... who pursue studies'); the verifier judged the substance unchanged. This licenses Malaysia's visa-access FAIL, which excludes the Forest City candidate — the consequence the holdback existed to gate. Reversal path if the structural bar is later shown wrong (an unlisted route granting STUDY+WORK_REMOTE): drop these enumerations back to NEEDS_EVIDENCE. | seed | seed | ec91ab4b-db89-4066-b3a2-88baa0770b3e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 82 | 178f0ed6-db96-402d-9771-d68cbda98001 | my | visa-access | {"passportSet":"CN","routesEnumerated":true} | NULL | https://web.archive.org/web/20260210001748/https://www.imi.gov.my/index.php/en/main-services/pass/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia), read via Internet Archive rendering | 2026-02-10 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Identical basis, closure argument, and disclosures to the US enumeration claim — four documents jointly close the authority's route list (IMI Pass index quoted verbatim there; IMI Visa Requirement by Country; MOTAC MM2H portal; MDEC DE Rantau FAQ V8), and IMI's Student Pass page makes the Student Pass the only study route, which carries no remote-work right. CN-SPECIFIC ADDITIONS: (a) the China–Malaysia Mutual Visa Exemption Agreement (in force 2025-07-17) is a CN-keyed entry route and is claimed separately with the PRC embassy's explicit statement that “Residence, employment, study, media activities, or other activities are not covered by the Agreement” — it adds tourist access, not admission; (b) IMI's menu carries an ‘eNTRI for Chinese PRC’ entry and a ‘FAQ Malaysia – China Mutual Visa Exemption’ page, neither readable this pass (imi.gov.my WAF-blocks this client) — both are entry-facilitation surfaces for short visits, named here so the record shows they were seen. Same tier disclosure: live imi.gov.my WAF-blocks this client IP; URL read is the Internet Archive rendering. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: Same four-document basis re-fetched and verified (see claimIndex 9, including the closure judgment and the two unnamed menu surfaces). Both CN-specific menu entries are present on my re-fetch of the capture; the separately-claimed mutual exemption (claimIndex 6) is verified with the embassy's own exclusion sentence, so the CN-specific statement that it 'adds tourist access, not admission' is the source's own words, not analysis. Enumeration holds for this set on the same nationality-neutral list basis. HELD AT NEEDS_EVIDENCE BY THE MERGE, 2026-07-18 — deliberately NOT pass-eligible, so this enumeration cannot license a FAIL yet. Both the researcher and the adversarial verifier flagged the same gap: the enumeration holds as a LIST, but the Entry Permit (PR) — a listed, in-principle qualifier — was never researched for its permits, and the verifier surfaced two unread menu entries ('FAQ – Graduate (Social) Visit Pass' and a work-permission-endorsement procedure). D-036's rule is that you cannot fail a cohort without evidence you read ALL of it, and 'all of it' includes each listed route's permits: a bar computed while a listed route's permits are unknown converts disclosed-not-proven into FAIL. UPGRADE PATH: read the Entry Permit conditions and the two menu entries; if none yields a qualifying route, raise this to CORROBORATED and the gate's FAIL becomes licensed. UPGRADED TO CORROBORATED BY THE MERGE, 2026-07-18 — the holdback is discharged. The three listed routes that were unread when this was held (Entry Permit/PR, Graduate (Social) Visit Pass, LTSVP-spouse + work-permission endorsement) have now been read from primary sources and adversarially verified (2 CONFIRMED, 1 CORRECTED, 0 REFUTED; evidence/visa-my/followup-enum-*). None permits RESIDE+STUDY+WORK_REMOTE in one route: Entry Permit lacks STUDY (Immigration Act 1959/63 silent on study for PR holders; the verifier independently confirmed no primary source grants or denies it, so silence is read conservatively as no grant), the Graduate pass lacks WORK_REMOTE, the spouse LTSVP lacks STUDY. THE COMPLETENESS DETERMINATION, made explicitly: the FAIL rests on a STRUCTURAL bar, not merely a compiled list — STUDY in Malaysia runs only through the Student Pass, which carries no remote-work right, so no single instrument can combine study with remote work regardless of list-completeness. RESIDUAL DISCLOSED: the Student-Pass-monopoly sentence the researcher quoted is no longer verbatim on the live IMI page (it now reads 'granted to non-citizen students ... who pursue studies'); the verifier judged the substance unchanged. This licenses Malaysia's visa-access FAIL, which excludes the Forest City candidate — the consequence the holdback existed to gate. Reversal path if the structural bar is later shown wrong (an unlisted route granting STUDY+WORK_REMOTE): drop these enumerations back to NEEDS_EVIDENCE. | seed | seed | 7d68b7c8-6782-4b82-a87a-18cb544b404f | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 83 | 97e63ceb-a291-44c7-a565-842d3c9e80c3 | my | visa-access | {"route":"Entry Permit (Permanent Residence) — Immigration Act 1959/63, s.10","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":null,"passportSet":"*"} | NULL | https://lom.agc.gov.my/ilims/upload/portal/akta/LOM/EN/Act%20155.pdf | Attorney General's Chambers of Malaysia — Laws of Malaysia, Act 155 (Immigration Act 1959/63), reprint incorporating all amendments up to 1 January 2006 | 2006-01-01 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | READS THE LISTED-BUT-UNRESEARCHED ROUTE THAT HELD THE ENUMERATION. The Entry Permit is the statutory instrument for permanent residence (PR): the Immigration Department's own definition is that an Entry Permit 'enable[s] the holder to apply for the Permanent Resident status'. SOURCE QUOTES, verbatim from the AGC statute text (Act 155): s.2 interpretation — “‘Entry Permit’ means a Permit to enter and remain in Malaysia issued under section 10”. s.10(1) — “Any person seeking to enter Malaysia who is not entitled so to enter as a citizen or by virtue of a valid Pass to enter Malaysia issued to him or seeking to remain in Malaysia after the expiry of such a Pass may make application in that behalf in the manner prescribed to the Director General…”. s.10(2) — “Upon application made under subsection (1) and upon payment of the prescribed fees the Director General may issue to the applicant an Entry Permit on such terms and conditions as the Director General may think fit in the prescribed form…”. s.55B(1) — “Any person who employs one or more persons, other than a citizen or a holder of an Entry Permit who is not in possession of a valid Pass shall [commit an offence]… not more than fifty thousand ringgit or to imprisonment for a term not exceeding twelve months or to both for each such employee.” WHAT IT PERMITS: RESIDE — s.10 grants a permit ‘to enter and remain in Malaysia’, i.e. residence; a PR Entry Permit does not carry a fixed per-entry expiry (it is the indefinite/permanent residence authorisation itself). WORK_REMOTE — s.55B makes employing a non-citizen without a valid Pass an offence BUT expressly exempts ‘a holder of an Entry Permit’, placing PR holders on the same footing as citizens for employment; a PR therefore has an unrestricted right to work in Malaysia, of which remote work for a foreign employer is a subset. WHAT IT DOES NOT PERMIT (as sourced): STUDY — the Act is silent on study, and IMI's Student Pass page states ‘All foreigners who wish to pursue their education in any level of education in Malaysia are required to apply for a Student Pass’; a PR is a non-citizen (foreigner) and is NOT exempted on the face of that statement, so on the sources the Entry Permit does not carry an as-of-right STUDY right in one route. (In administrative practice PRs are commonly treated as residents for education, but I found no citable instrument exempting PR holders from the Student Pass requirement, so STUDY is not recorded — silence is not permission.) maxStayDays=null RECORDS ‘INDEFINITE’, NOT ‘UNKNOWN’: PR residence has no fixed per-entry cap; s.10(2) issues the permit ‘on such terms and conditions as the Director General may think fit’ with no statutory duration, and a PR Entry Permit does not expire. Stay is therefore not the limiting factor; the route fails the scenario on STUDY (and on eligibility, below), so no number is needed and none is invented. ELIGIBILITY (for the note, distinct from permits — the distinction the last pass flagged): the Entry Permit is discretionary (‘as the Director General may think fit’), granted only on qualifying GROUNDS not open to a newly-arriving member — long prior lawful residence on other passes, marriage to a Malaysian citizen, a high-net-worth/expert/points basis, etc. — with a Ministerial appeal on refusal (s.10(3)). It is NOT an as-of-right route a scenario member could obtain on arrival; the gate owner should read this route as permitting RESIDE+WORK for those who already hold PR, not as a channel that admits the cohort. Because it does not carry STUDY on the sources, it does not create the single RESIDE+STUDY+WORK_REMOTE route the scenario needs, and the enumeration's closure argument survives it. PASSPORTSET ‘*’ BASIS (WEAKER FORM): s.10 states no nationality condition — the DG may issue to ‘any person’ on discretionary terms — so the route is nationality-neutral on the statute; I did not find an affirmative nationality-neutrality statement, and Malaysia's general Ministry-of-Home-Affairs posture toward Israeli nationals (a §5 passport-exclusion matter, not read by the gate) is not addressed by s.10. Recorded ‘*’ with that limit stated so it is not over-read. FETCH FACT (rendering): fetch_page of the AGC PDF triggers a browser download (NAVIGATION_FAILED, ‘Download is starting’); the AGC page-detail view (https://lom.agc.gov.my/act-detail.php?act=155&lang=BI) rendered live and confirms Act 155's identity and its Part III ‘ENTRY PERMITS’ / s.10 structure, and the full statute text was read via the r.jina.ai text extraction of this exact AGC PDF URL. EFFECTIVEAT is the reprint consolidation date (‘up to 1 January 2006’); the AGC timeline shows later subsidiary legislation and a Mar 2026 amendment, but the read text is the consolidated reprint and ss.2/10/55B are long-standing structural provisions — disclosed rather than assumed unchanged. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument (both agents WAF-walled identically, read via r.jina.ai proxy — a shared fetch fact). VERIFIER: fetch_page of the AGC PDF (lom.agc.gov.my/.../Act%20155.pdf) failed under my own instrument with NAVIGATION_FAILED ('Download is starting') — the same download behaviour the researcher reported. I independently read the full statute text via r.jina.ai text extraction of the exact AGC PDF URL (100,811 chars, status 200). All four cited quotes (s.2, s.10(1), s.10(2), s.55B(1)) verified verbatim against that text. RESIDE grounded in s.10 'enter and remain'; WORK_REMOTE defensibly grounded in the s.55B employer-offence exemption of 'a holder of an Entry Permit' (PR on citizen footing for employment). STUDY correctly withheld: the Act is silent on study (only 'educational' appears in the general Ministerial-restriction power, not as a study grant), and IMI's Student Pass page frames the Student Pass as the instrument 'granted to non-citizen students... who pursue studies in Malaysia' with no PR carve-out; no primary source grants study to an Entry Permit holder as of right, so the declination is the conservative reading of silence and is correct. maxStayDays null (indefinite) consistent with s.10(2) issuing on discretionary terms with no statutory duration. Claim value asserts no bar/enumeration. | seed | seed | a5fc6be0-d090-48d8-bf43-207880c4bdb2 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 84 | cbd5142c-36e8-4b73-95ea-14212c953c86 | my | visa-access | {"permits":["RESIDE","STUDY"],"maxStayDays":365,"passportSet":"*"} | NULL | https://www.imi.gov.my/index.php/en/contact-us/frequently-ask-question/faq-graduate-social-visit-pass/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia) | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | READS THE SECOND SURFACE THE VERIFIER SURFACED (‘FAQ – Graduate (Social) Visit Pass’). SOURCE QUOTES, verbatim from IMI's FAQ page: “This pass is a Long-Term Social Visit Pass that allows graduates to travel, continue their studies, and work part-time in permitted sectors”. Eligibility — “It is open to international students who have completed their studies at the Bachelor’s Degree level and above at Higher Education Institutions (IPT) in Malaysia”. Work scope — “work part-time in permitted employment sectors in accordance with currently enforced acts and laws”. Duration — “validity period?** 12 months”. And two closing negatives on the same page: it cannot be “extended” (‘No’) and its holder cannot use it to “apply for a Resident Pass/Entry Permit” (‘No’). WHAT IT PERMITS: RESIDE (a Long-Term Social Visit Pass, 12 months) + STUDY (‘continue their studies’). WHAT IT DOES NOT PERMIT: WORK_REMOTE — the only work right is ‘part-time in permitted employment sectors’, i.e. limited on-site work under Malaysian law, the same shape as the Student Pass's part-time allowance; it is NOT full-time remote work for a foreign employer. So this pass permits RESIDE+STUDY but not the scenario's WORK_REMOTE — it fails the all-three test on WORK_REMOTE. maxStayDays=365: the page's own ‘validity period 12 months’, and it is expressly non-extendable, so 365 is a hard per-pass ceiling (no renewal to count). ELIGIBILITY (distinct from permits): this is a POST-GRADUATION transition pass — only for a foreigner who has ALREADY completed a Bachelor's degree or higher at a Malaysian higher-education institution. It is unavailable to a newly-arriving scenario member, who has not studied in Malaysia; a member would first need a Student Pass (which itself lacks WORK_REMOTE). PASSPORTSET ‘*’ BASIS (WEAKER FORM): the page's stated scope is ‘international students’ of any nationality who graduated from a Malaysian IPT; no nationality condition and no affirmative neutrality statement — recorded ‘*’ with that limit so it is not over-read. FETCH FACT: live imi.gov.my WAF-blocks fetch_page against this client IP (HTTP 500, ‘Web Page Blocked!’, Attack ID 20000051, every path — same block both prior agents hit); this page was read via the r.jina.ai text extraction of the exact live IMI URL, which returns from a different server IP and is not WAF-blocked. A verifier can re-read the same way or by hand. effectiveAt is the retrieval date (the page states no rule-effective date). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CORRECTED, value above is the corrected one (passportSet narrowed from '*' to the page's enumerated nationalities; immaterial to the gate, the route fails on WORK_REMOTE for every set). VERIFIER: Live imi.gov.my WAF-blocked my own fetch_page (HTTP 500, 'Web Page Blocked!', Attack ID 20000051) — I hit the identical block. Read via r.jina.ai extraction of the exact IMI URL (status 200, 54,087 chars). permits [RESIDE, STUDY] CONFIRMED: RESIDE (Long-Term Social Visit Pass, 12 months) + STUDY ('continue their studies'); WORK_REMOTE correctly NOT attached (the only work right is 'part-time in permitted employment sectors', i.e. limited on-site work, not full-time remote work for a foreign employer). maxStayDays 365 CONFIRMED (validity 12 months, expressly non-extendable). CORRECTION: passportSet '*' is contradicted by the page, which explicitly answers 'Which nationalities are eligible? It is open to citizens of 32 countries' and enumerates them — the route is NOT nationality-neutral and '*' over-admits. Corrected to the 32 named nationalities. This does not change the qualifying outcome: the route fails the all-three test on WORK_REMOTE for every passport. Claim value asserts no bar/enumeration. CORRECTING QUOTE: “"This pass is a Long-Term Social Visit Pass that allows graduates to travel, continue their studies, and work part-time in permitted sectors"; "Which nationalities are eligible? It is open to citizens of 32 countries:" [lists 32]; "Working part-time in permitted employment sectors in accordance with currently enforced acts and laws"; "What is the validity period? 12 months"; "Can the pass be extended? No."” F7 RE-KEY (audit 2026-07-20, owner ruling): passportSet restored to '*'. The verifier's 32-nation value was stored as a single comma-joined string, which matches NO scenario cohort key and is itself a coined list — a D-037 breach (a bloc must be a quotation, not a coined key). '*' is the weaker form D-037 sanctions; the 32 nationalities remain quoted verbatim above, so no source claim is lost. Safe here specifically because the route fails WORK_REMOTE for every passport, so '*' cannot over-admit — the over-admit the verifier corrected against requires a QUALIFYING route, which this is not. If this route ever gained WORK_REMOTE, re-key to the tracked cohorts (per-state, like uz.json's EU treatment), never to a coined comma-string. | seed | seed | ef26464d-014a-4db8-95e3-57216c07856a | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 85 | ca0c6bd8-d444-4231-ac3d-8c12468080aa | my | visa-access | {"route":"Long Term Social Visit Pass (spouse of a Malaysian citizen) with Work Permission Endorsement","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":180,"passportSet":"*"} | NULL | https://www.imi.gov.my/index.php/en/terms-and-conditions-for-the-employment-of-a-citizens-spouse/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia) | 2026-07-18 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | READS THE THIRD SURFACE THE VERIFIER SURFACED (the ‘PROCEDURE FOR APPLYING FOR WORK PERMISSION ENDORSEMENT’). WHAT THE ENDORSEMENT IS AND WHOM IT COVERS — SOURCE QUOTES: the work-permission-endorsement page (https://www.imi.gov.my/index.php/en/main-services/pass/procedure-for-applying-for-work-permission-endorsement/) states “Applications can be submitted while obtaining the approval of the relevant Social Visit Pass or at any time after the relevant Social Visit Pass is issued.” and links to ‘Terms and conditions for the employment of a Citizen's Spouse’. That terms page (this sourceUrl) states the applicant “has already been issued such a pass as a foreign husband or wife to a Malaysian”, and that the “applicant has fulfilled the requirements specified by the Director General of Immigration, Malaysia for the Social Visit Pass to him/her as a husband or wife of a Malaysian and for the purpose of granting him/her permission to be engaged in paid employment or business or a professional occupation”. The Long Term Social Visit Pass page (https://www.imi.gov.my/index.php/en/main-services/pass/visitor-pass/social-visit-pass/long-term-social-visit-pass/) states the LTSVP is “for temporary residence in Malaysia for a minimum period of six (6) months” and that the endorsed spouse may “undertake professional employment without the need to convert to an Employment Pass or Visit Pass (Temporary Employment).” ANSWER TO THE QUESTION THE VERIFIER POSED — does this endorsement turn a social/visit pass into a scenario-relevant work route? NO. It only work-enables a Long Term Social Visit Pass that is itself issued on a MALAYSIAN FAMILY TIE: the LTSVP's own categories on that page are ‘foreign husband or wife of a Malaysian citizen’, ‘Children / Stepchildren to Citizens / Children whose citizenship status has not been determined’, and elderly parents sponsored by their Malaysian children (‘Sponsors must consist of children / daughters-in-law / siblings’, ‘Sponsors must have an income of RM2000 and above’). WHAT IT PERMITS: RESIDE (LTSVP, min 6 months) + WORK_REMOTE (the endorsement grants ‘paid employment or business or a professional occupation’ — an unrestricted work right of which remote work for a foreign employer is a subset). WHAT IT DOES NOT PERMIT: STUDY — neither the endorsement nor the LTSVP grants a study right (STUDY remains routed through the Student Pass per IMI). So this route permits RESIDE+WORK but not all three; it fails the scenario on STUDY. maxStayDays=180 is the SOURCED FLOOR, not a ceiling: the page states a ‘minimum period of six (6) months’ (≈180 days) and no maximum; recorded conservatively as 180 with this caveat, and it does not affect the verdict because the route fails on STUDY regardless. ELIGIBILITY (distinct from permits): requires marriage to a Malaysian citizen (‘marriage has been legally and properly registered under the laws in force in Malaysia’) or an equivalent Malaysian family sponsor — NOT available to a newly-arriving scenario member with no Malaysian family tie. PASSPORTSET ‘*’ BASIS (WEAKER FORM): nationality-neutral on its face (any nationality may marry a Malaysian) but conditioned on the family tie, not on nationality; no affirmative neutrality statement — recorded ‘*’ with the eligibility condition stated. FETCH FACT: live imi.gov.my WAF-blocks fetch_page against this client IP (Attack ID 20000051); all three IMI pages here were read via the r.jina.ai text extraction of the exact live IMI URLs (different server IP, not WAF-blocked). effectiveAt is the retrieval date (pages state no rule-effective date). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument (both agents WAF-walled identically, read via r.jina.ai proxy — a shared fetch fact). VERIFIER: Live imi.gov.my WAF-blocked my own fetch_page (HTTP 500, Attack ID 20000051); read via r.jina.ai extraction of the exact IMI terms URL (status 200, 51,901 chars). permits [RESIDE, WORK_REMOTE] CONFIRMED: RESIDE (LTSVP as foreign husband/wife of a Malaysian) + WORK ('paid employment or business or a professional occupation' endorsed by the DG — a general work right of which remote work for a foreign employer is a subset). STUDY correctly NOT attached (neither the endorsement nor the LTSVP grants study). CAVEAT: maxStayDays 180 is NOT stated on this sourceUrl (the terms page carries no duration); it is cross-sourced from the linked Long Term Social Visit Pass page's 'minimum period of six (6) months' and recorded as a conservative floor, as the note discloses — it does not affect the verdict, which turns on the absence of STUDY. passportSet '*' defensible: the page states no nationality condition (eligibility turns on a Malaysian family tie, not nationality). Claim value asserts no bar/enumeration. | seed | seed | 07f4aef2-5efd-4023-8fd4-bdc11cc6b6a5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 86 | d5490115-20f1-4d6b-a40e-d4b09736e7be | my | operating-licence | true | NULL | https://lom.agc.gov.my/ilims/upload/portal/akta/LOM/EN/Act%20555%20-%20Private%20Higher%20Educational%20Institutions%20Act%201996%20(As%20at%201%20December%202015).pdf | Attorney General's Chambers of Malaysia, Laws of Malaysia Act 555 (Private Higher Educational Institutions Act 1996) | 2015-12-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | UPGRADES THE SEED from NEEDS_EVIDENCE to true. The brief expected UNRESOLVED absent an actual licensing regime — the regime exists and is explicit in statute. Act 555 Part IV is titled 'ESTABLISHMENT OF A PRIVATE HIGHER EDUCATIONAL INSTITUTION WITH THE STATUS OF A UNIVERSITY, UNIVERSITY COLLEGE AND BRANCH CAMPUS', and s.21 (quoted verbatim below) expressly contemplates 'a branch campus of a foreign University or University College' subject to the Minister's prior approval. s.23 confirms the whole Act applies to such a campus. The old s.22 gatekeeper — which required an invitation from the Minister before a foreign institution could even apply — was deleted by Act A1465, liberalising entry. So: a foreign entity may lawfully run an educational institution in Malaysia, subject to a discretionary ministerial licence administered by MOHE/JPT. Extracted with pdftotext from the AGC gazette PDF (WebFetch could not read the binary). TWO CAVEATS the analyst must carry: (a) approval is discretionary, so 'lawful in principle' is not 'will be licensed'; (b) Act 555 governs HIGHER EDUCATION institutions, and whether Network School's programmes fall inside Act 555 at all is the precise question the Higher Education Ministry was directed to answer in the 2026-07-14 probe (see lawfulness). true here means the statutory door exists, not that this scenario is through it. Single-source, hence UNVERIFIED. SOURCE QUOTE, verbatim from the page on the retrieval date: “21. The prior approval of the Minister shall be obtained for the establishment of a private higher educational institution with the status of a University or University College or a branch campus thereof or a branch campus of a foreign University or University College.” | seed | seed | a705358a-921e-40a9-aac3-05be274a2110 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 87 | dab9f418-3b7e-4639-b216-73fd2ccd8be6 | kz-almaty-city | international-air-service | 1 | service (0/1) | https://ourairports.com/airports/UAAA/ | OurAirports (community open-data airport registry) | 2022-09-30 | 2026-07-20T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | D-047: the binary D-044 actually licensed, recorded on the criterion that can carry it. This row does NOT assert a route count — `flight-connectivity` stays empty for Almaty because no route count was obtainable. VERIFIER: CONFIRMED on value (evidence/almaty-city/verifier-verdicts.md §4); the shape defect the verifier flagged is what D-047 fixes. VERBATIM from the OurAirports facility-data panel (fetch_page, HTTP 200, 2026-07-20, every field re-fetched and CONFIRMED by the verifier): 'Name: Almaty International Airport', 'Location: Almaty, Almaty City, KAZAKHSTAN', 'IATA code: ALA', 'ICAO code: UAAA', 'Facility type: large_airport', 'Airline service?: yes', 'Coordinates: 43.354267,77.042828', 'Field elevation: 2,234 ft / 681 m MSL', 'Last updated: 2022-09-30 12:00:47'. 'Airline service? yes' is OurAirports' rendering of the CSV column scheduled_service="yes"; 'Facility type: large_airport' is the CSV type column. TIER NOTE: OurAirports is a community-maintained public-domain registry, not a civil-aviation-authority primary (that would be the Civil Aviation Committee of the Republic of Kazakhstan / AIP KZ). STALENESS CAVEAT, and it is worse here than for Bali or Dubai: this record's own 'Last updated' is 2022-09-30 — nearly four years before retrieval, against 2025-02-21 for Denpasar and 2026-02-15 for Dubai. The binary is almost certainly still true; the record backing it is the oldest of the four, the freshness decay reflects that, and the verifier said so explicitly. effectiveAt = the record's own 'Last updated' date. | seed | seed | 2885a96b-6703-4fd8-bbee-d211949725dd | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 88 | 079bbe8e-c78c-4961-8649-e68292f5e7b9 | my | foreign-title-statute | true | NULL | https://lom.agc.gov.my/ilims/upload/portal/akta/outputaktap/20201015_828_BI_WJW015175%20National%20Land%20Code%20Act%20828%20(Pewartaan).pdf | Attorney General's Chambers of Malaysia, Laws of Malaysia Act 828 (National Land Code, Revised 2020) | 2020-10-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | National statute permits it, conditionally. NLC Part Thirty-Three (A) ('RESTRICTIONS IN RESPECT OF NON-CITIZENS AND FOREIGN COMPANIES') s.433b(1) provides that a non-citizen or foreign company MAY acquire land and may take dealings in alienated land, 'but only after the prior approval of the State Authority has been obtained upon an application in writing'. s.433b(2): approval 'may be made subject to such terms and conditions as may be specified by the State Authority and to the payment of such levy as may be prescribed'. s.433a defines 'foreign company' to include a Malaysian-incorporated company with 50% or more voting shares held by non-citizens — so a locally-incorporated NS vehicle does not escape s.433b if foreign-controlled. Answer to the gate as posed ('as a matter of national statute') is therefore true. Extracted with pdftotext from the gazetted AGC PDF. CAVEATS: (a) consent is per-State and discretionary — Johor's consent is a live variable given the 2026-07-14 probe, in which the MB specifically flagged 'violations of licence or land conditions'; (b) state-level minimum purchase prices and levies are set by state rules I did not fetch and are out of scope for this pass; (c) s.433b(3) exempts charges/liens only. Single-source, hence UNVERIFIED. SOURCE QUOTE, verbatim from the page on the retrieval date: “but only after the prior approval of the State Authority has been obtained upon an application in writing to the State Authority by such non-citizen or foreign company:” | seed | seed | 603590f8-c47b-4a58-8420-927f996e8be5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 89 | 3c8fae64-ea9e-4471-a35b-313e2ffb6a3a | my | political-stability | 72.4670498 | WGI percentile rank (0-100, higher better) | https://api.worldbank.org/v2/country/MYS/indicator/GOV_WGI_PV.SC?format=json&source=3&date=2022:2024 | World Bank, Worldwide Governance Indicators | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | WGI 'Political Stability - Governance score (0-100)' for MYS, reference year 2024 (latest available; WGI source dataset lastupdated 2026-03-18). Trend is mildly improving: 2022 = 70.14, 2023 = 69.79, 2024 = 72.47. NOTE FOR THE PIPELINE: the legacy WDI indicator codes (PV.PER.RNK, PV.EST) now return 'The indicator was not found. It may have been deleted or archived.' — WGI has moved to source=3 with GOV_WGI_* codes. Any existing collector using the old codes is silently broken. Quote is the verbatim JSON object returned by the API. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator":{"id":"GOV_WGI_PV.SC","value":"Political Stability - Governance score (0-100)"},"country":{"id":"MY","value":"Malaysia"},"countryiso3code":"MYS","date":"2024","value":72.4670498,"unit":"","obs_status":"","decimal":1}” | seed | seed | 9d16872c-6026-49d5-9087-088bb1c9f180 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 90 | 91403f0b-af67-43c5-b013-b68694e1b591 | my | rule-of-law | 62.9047385 | WGI percentile rank (0-100, higher better) | https://api.worldbank.org/v2/country/MYS/indicator/GOV_WGI_RL.SC?format=json&source=3&date=2022:2024 | World Bank, Worldwide Governance Indicators | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | WGI 'Rule of Law - Governance score (0-100)' for MYS, reference year 2024 (latest available; dataset lastupdated 2026-03-18). Essentially flat: 2022 = 62.05, 2023 = 62.16, 2024 = 62.90. Same broken-legacy-code caveat as political-stability (RL.PER.RNK / RL.EST now 404 at the API; use source=3 + GOV_WGI_RL.SC). Quote is the verbatim JSON object returned by the API. Directionally consistent with Heritage's 2026 narrative for Malaysia ('The overall rule of law is relatively well respected in Malaysia'), which is why I marked this CORROBORATED rather than UNVERIFIED. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator":{"id":"GOV_WGI_RL.SC","value":"Rule of Law - Governance score (0-100)"},"country":{"id":"MY","value":"Malaysia"},"countryiso3code":"MYS","date":"2024","value":62.9047385,"unit":"","obs_status":"","decimal":1}” | seed | seed | a6b17e90-e325-44d6-bc02-0fa2b83f5c1c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 91 | ed089078-5499-4e1e-844d-acdfed2f2375 | my | corruption | 52 | Transparency International CPI score (0-100, higher better) | https://files.transparencycdn.org/images/CPI-2025-Report-EN.pdf | Transparency International | 2025-12-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | CPI 2025, Malaysia = 52/100, rank 54 of 182. Extracted with pdftotext from TI's own report PDF — the country score table is laid out score-first, so the verbatim row reads '52 Malaysia' (adjacent rows: Italy 53, Poland 53, Oman 52, Bahrain 50, China 43). Up 2 points from 50 in CPI 2024. Corroborated by Bernama's CPI wire report (https://cpi.bernama.com/news.php?id=2522403, titled 'CPI 2025: Malaysia's Ranking Improves To 54th, Score Rises To 52') and by TI Malaysia's own press release, both consistent on 52/54th. The transparency.org country landing page (https://www.transparency.org/en/cpi/2025/index/mys) does not render per-country scores to a fetcher — the report PDF is the reliable extraction path. SOURCE QUOTE, verbatim from the page on the retrieval date: “52 Malaysia” | seed | seed | d8f3d1f6-64ed-4760-b1b2-1e6514a29cec | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 92 | 79d988a7-5cec-4fb1-be18-f3523e2bf442 | my | expropriation-risk | 62.7 | Heritage Index of Economic Freedom - Property Rights score (0-100, HIGHER IS BETTER) | https://static.heritage.org/index/pdf/2026/2026_indexofeconomicfreedom_malaysia.pdf | The Heritage Foundation, 2026 Index of Economic Freedom | 2026-01-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | DIRECTION CHECKED: this is a property-RIGHTS score, higher = better = safer, so 62.7 is a middling-positive reading, not a 62.7% risk. From the 2026 edition's 'RULE OF LAW' chart on the Malaysia country page: the three chart values are Property Rights 62.7, Judicial Effectiveness 63.4, Government Integrity 52.9 (quote below is the verbatim chart row followed by its label row, as laid out by pdftotext). Malaysia's overall 2026 score is 68, world rank 45, 'MODERATELY FREE', up 0.9 from 2025; the page's own prose says 'The country's property rights score is above the world average'. Heritage's 2025 edition put property rights at 66.3, so this is a 3.6-point deterioration year on year — worth flagging. Extraction caveat: the numbers live in a chart, not a data table, so pdftotext recovers them positionally; I have read the label row and the value row and they align left-to-right. economicfreedom.heritage.org itself is behind Cloudflare (403 to both WebFetch and curl), so the static PDF is the only fetchable path — Heritage is NOT dead, just bot-walled. Heritage and Fraser are both licence-clean; I did not touch WJP, ICSID or UNCTAD. SOURCE QUOTE, verbatim from the page on the retrieval date: “62.7 63.4 52.9 0 0 Property Judicial Government Rights Effectiveness Integrity” | seed | seed | d48445a8-6787-4370-956f-e6a3590982e5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 93 | be636f43-8c6b-45da-9e81-391c71a21408 | kz-almaty-city | disaster-exposure | 2.8 | INFORM Risk 0-10 | https://data360api.worldbank.org/data360/data?DATABASE_ID=DRMKC_INFORM&REF_AREA=KAZ | World Bank Data360 (relaying JRC/DRMKC INFORM Risk Index) | 2025-01-01 | 2026-07-22T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | OWNER RULING (D-044): INFORM as a COARSE SCREEN, tier-flagged, and it is NATIONAL. VERIFICATION HISTORY, recorded because it is unusual: the Almaty verifier could NOT re-fetch this value — data360api.worldbank.org returned 'page.goto: Timeout 45000ms exceeded' on SEVEN attempts across three URL variants — so the claim stood UNVERIFIED (evidence/almaty-city/verifier-verdicts.md §2) and was explicitly barred from landing on the researcher's say-so. It was re-fetched independently on 2026-07-22 (this seeding pass, fetch_page, HTTP 200) and every figure in this note was confirmed: INDICATOR 'INFORM_OVRL', REF_AREA 'KAZ', UNIT_MEASURE '0_TO_10', 'TIME_PERIOD': '2025', 'LATEST_DATA': true carries "OBS_VALUE": "2.8". The three 2025 sub-scores this composite launders were each re-fetched separately and also confirmed: INFORM_HA (Hazard & Exposure) 3, INFORM_VU (Vulnerability) 2.2, INFORM_CC (Lack of Coping Capacity) 3.2, all LATEST_DATA true. The same call returns the decade series (2.4-2.5 across 2016-2022, 2.8 for 2023/2024/2025), so 2.8 is not a one-year artefact. COARSE-SCREEN CAVEAT (DATA-SOURCES.md §5): INFORM 'is already a score' and launders its sub-scores; this is a COUNTRY-level composite used as a coarse screen, NOT a reviewable CITY observation — attached to Almaty as INHERITED-NATIONAL. IT IS ACTIVELY MISLEADING FOR THIS CITY AND THAT MUST BE SAID: Kazakhstan is vast, mostly flat and mostly aseismic, and the national roll-up flattens the fact that Almaty sits in a documented high-seismicity zone. VERBATIM from https://en.wikipedia.org/wiki/Almaty (fetch_page, 2026-07-20, quotes re-checked by the verifier): 'Industrially developed and populated areas in the south and southeast of Kazakhstan are situated in the zones where the maximum magnitudes of expected earthquakes are from 6.0 to 8.3 (the intensity of I0=8-10)'; 'In each of these earthquakes, the city suffered wide destruction'; 'On 3 January 1911, the city was almost completely destroyed with over 770 brick buildings collapsing as a result of the 1911 Kebin earthquake.' That Wikipedia section self-flags as needing more citations and is cited here ONLY as a caveat on the national number, never as the value. A national 2.8 that reads 'low risk' is the coarse screen failing exactly where it always fails, and BOTH city verifiers independently concluded that seismic hazard has no home in the CITY criterion set — an open contract gap, not something this row covers. TIER NOTE: the URL read is the World Bank Data360 mirror (MULTILATERAL) of the primary JRC/DRMKC INFORM index. effectiveAt = the 2025 release year (start-of-year convention; the API record carries no exact release date). retrievedAt = 2026-07-22, the confirming fetch. | seed | seed | 0ff5f4c4-dce3-46c9-b563-d895ec5e3e23 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 94 | 54fa4039-a4e7-4ef8-95a4-cc6e5c18624a | kz-almaty-city | local-additionality | 1 | rubric 0-5 | https://adilet.zan.kz/rus/docs/P2300000349 | Government of the Republic of Kazakhstan, Resolution No. 349 of 3 May 2023 approving the General Plan of the city of Almaty (adilet.zan.kz, the state legal information system) | 2023-07-26 | 2026-07-22T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SCORE UNCHANGED AT 1, AND THE DEEP PASS TRIED HARD TO RAISE IT AND FAILED — that attempt is the strongest evidence for this score. A researcher proposed 3 on Government Resolution No. 785 of 13.09.2023 (the action plan for Almaty's socio-economic development), which funds 12,000 new student bed-spaces by end-2025 and 20,000 more by end-2030 and doubles the state subsidy (92 -> 184 MRP per bed-space) for converting non-purpose-built buildings into university dormitories. The adversarial verifier re-fetched it, confirmed all eleven quoted Russian fragments character by character INCLUDING the source's own missing space in '-152,3 млрд тенге', confirmed the instrument is in force (history shows one row, 'Статус НПА: Новый', and it appears unstamped in a search list where repealed siblings carry 'Утративший силу'), and still REJECTED the raise on the anchor's own verb. rubric.md §2 anchor 3 limb 2 requires a gap the node FILLS. That plan's gap is 'койко-места' in 'общежития вузов', delivered by 'Содействие вузам', paid by a госзаказ whose eligibility condition is that the building becomes a вуз dormitory. A Network School node is not a вуз and cannot contribute one bed to the 12,000 or the 20,000. An instrument that names a project class and EXCLUDES ours cannot lift the score — 'never infer upward' decides it. The verifier also inverted the calibration argument worth recording: a quantified, funded gap necessarily carries an eligibility test, while an unquantified activity list does not, so Almaty is the FIRST of the three cities to fall on strictness, not the last. Bali is not implicated (its PP 23/2023 names an activity to be hosted with no beneficiary-eligibility test); Tashkent is exposed to the same test but cannot be run against it, because its instrument is a press report of a proposal rather than a gazetted act — flagged for re-examination when it is gazetted. ANCHOR 1 SURVIVES ON BOTH CONJUNCTS: ~300 residents is ~0.013% of 2.37 m (statistical noise), and no local plan seeks this activity. THERE IS NO ROUTE TO 2: D-035's cap is a ceiling, not a grant, and anchor 2's district-level employment or business counts DO NOT EXIST — stat.gov.kz publishes housing commissioning, investment and average wage by district, but not employment or enterprise counts, re-verified this pass. The outcome space here is binary, 3 or 1, and it is 1. CITATION UPGRADED THIS PASS: the negative now rests on the PRIMARY. The Genplan-2040 approving instrument was located at adilet.zan.kz (Resolution No. 349 of 3 May 2023, which repeals the 2002 instrument and carries no repeal marker of its own), replacing the earlier citation of the almatygenplan.kz state-institute relay (OFFICIAL_SECONDARY). A full-text token search of the Genplan found ZERO occurrences of 'общежит', 'студен', 'вуз', 'университет' or 'высшего образования' — the plan names general K-12 schools, clinics, utilities, transit, green space and seismic retrofit, and no residential education/talent/knowledge-hub activity of this project's class. DATES, AND DO NOT CONFLATE THEM (verifier precision defect, applied): 2023-05-03 is the ADOPTION date; the resolution commences 'со дня его первого официального опубликования' (САПП No 19 art. 140, to press 26.07.2023), so effectiveAt is the commencement date 2023-07-26. SCOPE DISCIPLINE, CONFIRMED BOTH WAYS BY THE VERIFIER — two places, near-identical names: 'посёлок Алатау, Медеуский район' IS inside Almaty city (the Genplan names an IT park there), while 'город Алатау Алматинской области' is a separate settlement in Almaty REGION (Resolution No. 388 of 17.05.2024, with 'Алматинской области' on the instrument's own face) and is correctly NOT scored here. OPEN LEAD, AND IT POINTS DOWN NOT UP: a search summary the verifier deliberately declined to record suggests Almaty already operates ~136 dormitories at ~46,200 bed-places. If that is retrievable at a real source, the next honest move on this criterion is toward anchor 0 (saturation), not upward. STILL UNREAD: the Almaty Development Programme to 2030 (located, but the 4 MB PDF is download-walled) and a newer 'План развития города Алматы на 2026-2030 годы' (maslikhat, 10.12.2025, marked in force, paywalled at 800 tenge) — note a maslikhat decision cannot repeal a Government resolution, so neither displaces the finding above. | seed | seed | cb265d07-15e8-4614-859e-af7cbbfb6fa5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 95 | 4c6a2990-4ce6-4ca9-ad0e-dde019b2e98b | my | currency-inflation | UNRESOLVED | CPI %/yr | https://www.imf.org/external/datamapper/api/v1/PCPIPCH/MYS | International Monetary Fund, World Economic Outlook | 2025-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | NEEDS_EVIDENCE | OFFICIAL_API | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed 1.4 and an independent agent re-fetched the source and rejected it. VERIFIER: COULD NOT CHECK — not evidence the source is bad. The IMF DataMapper endpoint returned HTTP 403 Forbidden to WebFetch, and a direct curl with a browser User-Agent also returned 403 (Akamai 'Access Denied', reference #18.ae4d52cb). I never saw the payload, so I can neither confirm nor refute the claimed 2025 value of 1.4. Separately worth flagging for a re-check: the claimed quote is rendered in Python dict syntax with single quotes, which is not what a JSON API returns — that is a formatting tell that the quote may have been transcribed through a script rather than copied from the response, though it is not proof of anything. Needs a reachable source (e.g. IMF WEO database or BNM). ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: IMF WEO indicator PCPIPCH (Inflation rate, average consumer prices, annual % change) for MYS. Reference year 2025 = 1.4%. Series context: 2022 = 3.4, 2023 = 2.5, 2024 = 1.8, 2025 = 1.4, 2026 = 1.9 (projection), 2027 = 2.0 (projection). I report 2025 as the most recent non-projection year; note WEO 2025 values may still be estimates rather than final outturns. This is comfortably at the benign end of the 0-50 scale — inflation is not a risk axis for Malaysia. Quote is the verbatim subset of the JSON payload for MYS. Heritage's 2026 Malaysia page independently notes 'Monetary stability is relatively well maintained', consistent. QUOTE THE RESEARCHER OFFERED: “{'2022': 3.4, '2023': 2.5, '2024': 1.8, '2025': 1.4, '2026': 1.9, '2027': 2}” | seed | seed | 958650bc-f42f-4c31-a8ae-57b5c683a601 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 96 | 349dd098-041e-4987-8cf4-227e9130761f | my | english-proficiency | 581 | EF EPI score (300-700) | https://www.ef.com/wwen/epi/regions/asia/malaysia/ | EF Education First, English Proficiency Index | 2025-12-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | MODEL_ASSISTED_RESEARCH | NULL | EF EPI 2025 edition: Malaysia scores 581, global rank #24, 'High' proficiency band; global average 488, so Malaysia is +93. This is a genuine strength and one of the few axes where Malaysia clearly outperforms regional peers. Two caveats, both as the brief anticipates: (1) EF is the only source for this metric, so there is nothing to corroborate against — UNVERIFIED is structural, not a gap I can close; (2) EF EPI LICENSING FOR COMMERCIAL USE IS UNRESOLVED — EF is not on the prohibited list, but I found no affirmative commercial licence grant either, and the brief flags this. Do not treat the licence question as cleared. The edition year is inferred from the page's own fact-sheet link path ('.../2025/ef-epi-fact-sheet-malaysia-english.pdf'); EF does not label the edition prominently in the fetched body. SOURCE QUOTE, verbatim from the page on the retrieval date: “EF EPI score: 581” | seed | seed | 145b2281-8379-4d01-b244-9858a7863487 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 97 | 7e31d155-bc3f-48b5-b215-054ea6849967 | my | entry-friction | 0 | cohort-weighted days to obtain a visa (0-90, lower better) | https://www.imi.gov.my/index.php/en/main-services/visa/visa-requirement-by-country/ | Immigration Department of Malaysia (Jabatan Imigresen Malaysia) | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | MODEL_ASSISTED_RESEARCH | NULL | READ THE CAVEAT BEFORE USING THE 0. THE NUMBER: all four cohorts face zero pre-departure visa days on the ORDINARY entry route. Weighted: US 0 (x0.30) + EU 0 (x0.25) + IN 0 (x0.25) + CN 0 (x0.20) = 0. SOURCING, per cohort, from IMI's own list (quote below is the verbatim visa-required list): US — absent from the visa-required list, therefore not visa-required. EU — no EU member state appears on the list (Montenegro and Serbia are listed but are not EU); Netherlands, Ireland, Cyprus and Malta additionally appear on IMI's 'do not require a visa for any purpose of entry' table. IN — India IS on the visa-required list, but the page's own NOTE overrides it: 'India** citizen: visa exempts until 31st December 2026.' THIS EXPIRES IN ~5 MONTHS and IMI states no successor — the IN cohort's 0 has a hard date on it and must be re-checked. CN — absent from the list, and covered by the China-Malaysia Mutual Visa Exemption in force since 2025-07-17 (30 days/entry, 90 per 180). THE CAVEAT THAT MATTERS: 0 is true for tourists and false for anyone who could actually attend. The only route that gets near the scenario is DE Rantau, whose processing time I could NOT source officially — every figure I found ('6-8 weeks', 'really 4-6 months') traces to relocation agencies and nomad-visa blogs, which rule 8 excludes, and MDEC's page states no SLA. If the engine is weighting entry-friction against the gate route rather than ordinary entry, this 0 is the wrong denominator and the right answer is NEEDS_EVIDENCE. Flagged in blocked. Fetched by curl --compressed; WebFetch mis-summarised this page's tables (it wrongly reported the US as visa-required). SOURCE QUOTE, verbatim from the page on the retrieval date: “Countries required to apply for a visa to enter Malaysia (List of involved countries)|***Afghanistan (VDR)|Angola*|Bangladesh|Bhutan|Burkina Faso*|Burundi*|Cameroon*|Central African Republic*|Colombia|Democratic Republic of Congo*|Congo Republic*|Djibouti*|Equatorial Guinea*|Eritrea*|Ethiopia*|Ghana*|Guinea-Bissau*|India**|Ivory Coast (Cote D' Ivoire)*|Liberia*|Mali*|Montenegro|Mozambique*|Nepal|Nigeria*|Niger*|Pakistan|Rwanda*|Serbia|Sri Lanka| | |NOTE| |For countries marked as (*) are allowed to enter Malaysia by air only.|India** citizen: visa exempts until 31st December 2026.” | seed | seed | 7c24566b-d005-428f-ac8d-728a88413ee2 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 98 | e58b752f-25c3-4fa3-af78-6ee72d1942c1 | my | tax-burden | UNRESOLVED | USD/member-month | https://taxsummaries.pwc.com/malaysia/individual/income-determination | PwC Worldwide Tax Summaries | 2026-07-17 | 2026-07-17T00:00:00.000Z | COMMERCIAL_LISTING | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | I CAN GET A PLAUSIBLE NUMBER HERE VERY EASILY AND I AM DECLINING TO. The tempting answer is USD 0/member-month: Malaysia is broadly territorial, and foreign-sourced income received by RESIDENT individuals is exempt under the Income Tax (Exemption) (No. 5) Order 2022 [P.U.(A) 234/2022], reportedly extended from 31 Dec 2026 to 31 Dec 2036. Applied to DE Rantau's sourced USD 24,000/yr income floor that yields a clean, sourced-looking USD 0. It is not safe, for three independent reasons. (1) RESIDENCY: the exemption is for RESIDENT individuals; Malaysian tax residency needs 182 days. The scenario's floor is 180 days — members could land just SHORT of residency and outside the exemption entirely, where non-residents face a flat 30% on Malaysian-source income. The gate threshold and the residency threshold are two days apart and nobody has reconciled them. (2) SOURCE: employment/business income is generally sourced where the work is PERFORMED. A DE Rantau holder physically working from Malaysia has a serious argument that the income is Malaysian-sourced, not foreign-sourced — in which case the FSI exemption never engages, whatever their residency. I found no LHDN ruling, public determination or guideline addressing the source of a DE Rantau holder's income. (3) SOURCES: every asserted answer I found ('0% tax digital nomad visa') is a relocation firm, nomad-tax blog or Big-4 summary, and they contradict each other on the exemption end date (2026 vs 2036) and on the 60/182-day treatment. Rule 8 territory exactly. The primary instrument was not reachable: the LHDN FSI guidelines PDF at hasil.gov.my 404s at the URL surfaced by search, and I did not locate the gazetted extension order. sourceUrl records what I actually looked at and rejected. TO RESOLVE: fetch P.U.(A) 234/2022 and its extension order from lom.agc.gov.my, plus the current LHDN 'Tax Treatment In Relation To Income Received From Abroad' guideline, and get a Malaysian tax opinion on source-of-income for DE Rantau holders. Also note Heritage 2026 records 'The top individual income tax rate is 30 percent... The tax burden equals 13.1 percent of GDP' — macro context, not a per-member figure. SOURCE QUOTE, verbatim from the page on the retrieval date: “Foreign-sourced income received in Malaysia from outside Malaysia by resident individuals is subject to tax. However, the following income received in Malaysia from 1 January 2022 to 31 December 2036 may qualify for tax exemption, subject to conditions” | seed | seed | 700ceb10-b768-47b6-b41d-67282286f5ef | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 99 | f51f1b08-6917-4d3c-be80-fcce56f78241 | my | import-duty | UNRESOLVED | USD | https://www.trade.gov/country-commercial-guides/malaysia-import-tariffs | US Department of Commerce, International Trade Administration | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | No tariff line, no number. The criterion requires citing the specific tariff line used, and I could not read Malaysia's actual schedule for it. WHAT I HAVE: only the aggregate — ITA reports a 'simple average applied tariff of 6.1 percent for industrial goods' and SST on imported taxable goods at '5 or 10 percent, or a specific rate depending on the category of products'. An average is not a tariff line; applying 6.1% to an assumed equipment basket would be invention twice over (invented rate, invented basket). WHAT BLOCKED ME: JKDM's official HS Explorer (https://ezhs.customs.gov.my/) returns HTTP 200 but is a jQuery/DataTables SPA that renders tariff rows client-side; I probed for a JSON endpoint and found none exposed. Customs is live, not dead — it just needs a browser or the published tariff PDF. WHAT I DELIBERATELY DID NOT DO: quote a duty rate for HS 8471.30 (laptops) from dutyskip/credlix/dripcapital/flexport/tariffnumber — these are commercial HS-lookup aggregators, several were showing other countries' schedules, and none is Malaysia's customs authority. A wrong duty rate with a real-looking URL is precisely the failure mode this database exists to prevent. TWO LEADS WORTH FOLLOWING, both of which could move this to ~0: (a) Malaysia is an ITA signatory and computers under HS 8471 are widely duty-free — but this must be read off JKDM's schedule, not assumed; (b) Forest City sits inside a duty-free zone per the SFZ materials, which may zero the duty for this specific scenario — that is a zone-level fact and likely belongs to the city/property pass, not this one. TO RESOLVE: drive ezhs.customs.gov.my with a browser for HS 8471.30 / 8471.41, or pull the gazetted Customs Duties Order. SOURCE QUOTE, verbatim from the page on the retrieval date: “Malaysia's tariffs are typically imposed on a value-add basis, with a simple average applied tariff of 6.1 percent for industrial goods.” | seed | seed | 5bad5f61-50ff-4e6a-a8ec-0f75278a1bf3 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 100 | 9676b1ed-ae58-4f1e-b605-5f6922834106 | my | government-receptivity | 1 | rubric 0-5 | https://mediadigitaljohor.gov.my/en/kerajaan-johor-mahu-siasatan-menyeluruh-isu-network-school-di-forest-city/ | Kerajaan Negeri Johor (Johor State Government, Media Digital Johor) | 2026-07-14 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 1: No engagement and no applicable policy; FDI posture neutral or silent on this activity — reached downward from higher anchors because anchor 0's "OPENED AN INVESTIGATION INTO it" condition is met but no official has opposed the project CLASS as such THE FACTS: · On 14 July 2026 the Johor Menteri Besar, Onn Hafiz Ghazi, ordered a comprehensive investigation into Network School at Forest City, directing KDN, Immigration, police, Customs, the Higher Education Ministry and the Digital Ministry to check security, immigration and regulatory compliance. https://mediadigitaljohor.gov.my/en/kerajaan-johor-mahu-siasatan-menyeluruh-isu-network-school-di-forest-city/ “Kerajaan Negeri meminta Kementerian Dalam Negeri (KDN), Jabatan Imigresen Malaysia (JIM)...” · The federal Home Ministry opened a separate probe into Network School on 14 July 2026. https://www.freemalaysiatoday.com/category/nation/2026/07/14/network-school-in-forest-city-faces-home-ministry-probe “the ministry, through the immigration department, the police and other agencies, is conducting a comprehensive review and investigation” · The federal engagement that preceded the probes was closed-door and produced no instrument: Srinivasan pitched Digital Minister Gobind Singh Deo on a closer government relationship at private meetings on 17 April 2026. https://theedgemalaysia.com/node/803532 “According to people present at closed-door meetings last month, the former Coinbase Global Inc executive asked officials including Minister of Digital Gobind Singh Deo to consider a tighter-knit relationship between the government and his Network School, which he’s positioned as a means to bring world-class tech talent to the Southeast Asian nation.” · The one public ministerial endorsement of Network School was retracted — Gobind posted praise after his 17 April 2026 site visit, then deleted it. https://theedgemalaysia.com/node/803532 ““What I saw reaffirmed Malaysia’s position as a rising destination for global tech talent,” he wrote in a social media post shortly after the visit... Later, however, he deleted the post.” FACTS: Two agencies have open investigations into this exact project as of 2026-07-14 — the Johor MB ordered a whole-of-government probe (Johor state government, primary) and the Home Ministry opened its own (FMT). MAPPING: anchor 1. Anchor 0's own language is "OPENED AN INVESTIGATION INTO it," and that condition is literally met; the score is 1 rather than 0 only because the probes target immigration/identity compliance at this one site, not the project class as such — no official has opposed network-state/co-living tech communities as a category. WHAT ARGUED HIGHER AND WHY IT FAILED: Anchor 4 has a real case — Digital Minister Gobind Singh Deo MET about this specific project on 17 April 2026, visited the site, and posted public praise; Johor investments exco Lee Ting Han told Bloomberg the school brought "energy" to the area. Anchor 4's text ("has met... or issued a public statement about THIS SPECIFIC PROJECT... ON THE RECORD") is superficially satisfied. It does not reach because (a) The Edge confirms the meetings were closed-door — private meetings are not "on the record"; (b) Gobind deleted the post, so the only public endorsement was retracted and is not a standing statement; (c) no instrument exists — no MOU, licence, or gazetted incentive names this project or its class, which caps anchor 5 outright. Anchor 3 (named policy fit) and anchor 2 (generic openness) are also outranked: the Johor-Singapore SEZ/Forest City SFZ incentives are gazetted but name no project class, capping at 2. Against all of it, a live investigation dominates. TO MOVE UP: a gazetted incentive or signed MOU naming the project/class (5), or an on-the-record statement of support from an empowered official that is not subsequently retracted, after the probes close (4). TO MOVE DOWN to 0: any named official opposing the project class as such, or a revocation/restriction order issuing from the probes. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: All four facts re-fetched and hold verbatim. mediadigitaljohor.gov.my carries 'Kerajaan Negeri meminta Kementerian Dalam Negeri (KDN), Jabatan Imigresen Malaysia (JIM), Polis Diraja Malaysia (PDRM), Jabatan Kastam Diraja Malaysia (JKDM)...menjalankan semakan dan siasatan menyeluruh', dated 14 July 2026, MB Onn Hafiz Ghazi, into Network School at Forest City. FMT carries the Home Ministry probe quote at the same date. The Edge carries both the closed-door meeting passage naming Gobind Singh Deo and the 'What I saw reaffirmed Malaysia's position...' post plus 'Later, however, he deleted the post.' MAPPING: I pressed hard on anchor 0, which is the live question — a named official and a named agency have each literally 'OPENED AN INVESTIGATION INTO' this project, which is anchor 0's own language, and the note's site-vs-class distinction is not drawn in the anchor's condition list ('it or a close analogue', where 'it' is the project). But the rubric's WORKED EXAMPLE adjudicates this exact fact pattern by name: 'Malaysia is 1... It is 1 rather than 0 only because no official has opposed the project CLASS as such.' The contract resolves its own edge case and the note tracks it precisely. Anchor 4 correctly declined: The Edge establishes the meetings were closed-door (not 'ON THE RECORD') and the only public endorsement was retracted. Anchor 5 capped by absence of any instrument. | seed | seed | beeb676d-b190-4fda-9fe8-9023be959947 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 101 | ba142844-7bce-4f01-9362-6d42465ef807 | my | visa-preference | 4 | rubric 0-5 | https://www.mdec.my/derantau | Malaysia Digital Economy Corporation (MDEC), agency under the Ministry of Digital | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A preferential route exists, is in force, open AS OF RIGHT on published criteria, covers ALL FOUR cohorts THE FACTS: · The DE Rantau Nomad Pass is open to citizens of every nationality except Israel — so all four cohorts (US/EU/IN/CN) are covered and none is excluded, quota'd or restricted. https://www.mdec.my/derantau “The pass is open for application to citizens of all nationalities except Israel.” · It is a Professional Visit Pass granted for 3 to 12 months, renewable once for up to a further 12 months (24 months maximum), with published fees — i.e. in force and transacting, not pilot-stage. https://www.mdec.my/derantau “Type of Pass: Professional Visit Pass (Pas Lawatan Ikhtisas (PLIK)) Features: • Stay from 3 up to 12 months • Pass renewable for up to additional 12 months • Able to bring in spouse & child/children • Able to bring parent/parents for main pass holder only Fee: • MYR 1,000.00 (main) • MYR 500.00 (dependent)” · Eligibility runs on published objective criteria — profession domains plus income thresholds of USD 24,000 (tech) and USD 60,000 (non-tech) — and covers remote workers and independent contractors, the Network School cohort. https://www.mdec.my/derantau “Type of Digital Nomad Pass: Tech Talent/Profession and Non-Tech Talent/Profession are eligible to apply! Digital Freelancer, Independent Contractor Remote worker (full and part-time) i.e. employee of a company but not required to physically be in the office” FACTS: DE Rantau Nomad Pass (MDEC, live and accepting registrations as of retrieval) is a remote-work Professional Visit Pass, 3–12 months renewable once to 24 months, open to all nationalities except Israel, on published income/profession criteria with published fees. MAPPING: anchor 4 — a preferential route beyond tourist/business entry, in force, on published criteria, covering all four cohorts. WHAT ARGUED HIGHER AND WHY IT FAILED: Anchor 5 was genuinely arguable and is the closest call here. DE Rantau IS "designed for this activity class" in the sense the anchor contemplates — it is a remote-work route and Network School's members are remote workers who squarely fit it. It fails on the second conjunct: anchor 5 requires "published multi-year terms," and the published TERM of a grant is 3–12 months with a single discretionary renewal; a 24-month ceiling assembled from two sub-annual grants is not multi-year terms. It also fails on fit-of-the-project as opposed to fit-of-the-member: DE Rantau is built for individual nomads, not for a residential education institution, and the rubric warns that a route open to everyone equally is a 4 at most — nationality-agnosticism is not preferentiality. Note the MM2H SEZ/SFZ category (mm2h.gov.my) is tied to the very zone Forest City sits in, but it is an investor route — wrong activity for this cohort, which is anchor 2 for that route, below DE Rantau's 4. Not anchor 0: no restriction, quota or exclusion touches US/EU/IN/CN. TO MOVE DOWN to 3: evidence that MDEC/Immigration approval is in practice discretionary rather than criteria-driven — the MDEC page publishes criteria but nowhere disclaims discretion, and this is the main residual risk in the 4. TO MOVE UP to 5: a published multi-year remote-work or education route, or a route created for this project class. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: All three facts re-fetched at mdec.my/derantau and hold: 'open for application to citizens of all nationalities except Israel'; Professional Visit Pass, 3-12 months, renewable for an additional 12, fees MYR1,000/MYR500; tech USD24,000 / non-tech USD60,000 thresholds covering remote workers and independent contractors; actively accepting applications. MAPPING: anchor 4's conjuncts are each met — a route beyond ordinary tourist/business entry (so above anchor 1), in force and transacting (so not anchor 3's 'pilot-stage, or not in force'), published objective criteria, and no cohort among US/EU/IN/CN excluded (Israel is not a scored cohort, so anchor 2's 'a cohort excluded' does not bite; duration covers the 180d gate, so no 'wrong duration'). I specifically hunted the discretion risk the note flagged as its own residual weakness: the only evidence of discretion ('MDEC provides an approval recommendation', 'digital attributes remains quite subjective') traces to relocation agencies and nomad blogs, not to any official disclaimer, and cannot carry a REMAP to anchor 3 against a primary page publishing objective criteria. Anchor 5 correctly declined — the published grant term is sub-annual with one renewal, which is not 'published multi-year terms', and the rubric's own NOTE caps a nationality-agnostic route at 4. | seed | seed | dafc0d2c-e2f9-4c90-a94d-0abdfda78f70 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 102 | 2d63bb2b-d9b8-4acf-a822-f40c0ae3f5c3 | my | web3-support | 3 | rubric 0-5 | https://www.sc.com.my/regulation/guidelines/recognizedmarkets/list-of-registered-digital-asset-exchanges | Securities Commission Malaysia | 2026-06-26 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: Framework PLUS published tax treatment PLUS documented public-sector engagement THE FACTS: · Five Digital Asset Exchanges are registered and operating in Malaysia under the SC's Recognized Market Operator regime, per the SC's own list updated 26 June 2026: HATA Digital, Luno Malaysia, MX Global, SINEGY DAX and Kinetic DAX. https://www.sc.com.my/regulation/guidelines/recognizedmarkets/list-of-registered-digital-asset-exchanges “Updated list as of 26 June 2026” · BANKING LIMB (rubric requires this explicitly): banking is not merely obtainable but mandated — the SC's Guidelines on Recognized Markets para 15.28(h) requires every DAX operator to hold client monies in trust accounts at a licensed Malaysian financial institution, which the same Guidelines define as a licensed bank under the Financial Services Act 2013. https://www.sc.com.my/api/documentms/download.ashx?id=8dee8828-6793-42a8-95d9-074791d7a69c “establish and maintain in a licensed Malaysian financial institution one or more trust accounts, designated for the monies received from investors;” · Published tax treatment: the Ministry of Finance states that income from digital currency trading, mining and exchange is taxable under section 4 of the Income Tax Act 1967 (statement dated 29 August 2025); LHDN's Guidelines on Tax Treatment of Digital Currency Transaction (2nd ed., 5 December 2025) elaborate it. https://mof.gov.my/portal/ms/berita/akhbar/pendapatan-mata-wang-kripto-tertakluk-di-bawah-akta-cukai-pendapatan-1967-mof “Pendapatan hasil aktiviti urus niaga mata wang digital kripto seperti perdagangan, perlombongan dan pertukaran di Malaysia adalah tertakluk kepada cukai pendapatan di bawah Seksyen 4 Akta Cukai Pendapatan 1967.” · Documented public-sector engagement: the SC runs a named programme, aFINity, presented on its own digital-assets page, and revised the DAX guidelines on 20 May 2026 to strengthen client asset safeguards. https://www.sc.com.my/digital-assets “In September 2015, the Securities Commission Malaysia (SC) launched the 'Alliance of FinTech Community' (aFINity) to catalyse and nurture Fintech development in the Malaysian capital market.” REMAPPED BY ADVERSARIAL VERIFICATION (D-032/D-035). The scorer claimed 4 at "anchor 4: Framework PLUS published tax treatment PLUS documented public-sector engagement". An independent agent re-checked the facts against the pages AND the mapping against contracts/rubric.md, and the facts do not reach that anchor. VERIFIER: FACTS HOLD, all four. I extracted the SC Guidelines PDF myself rather than rely on the fetcher: para 15.28(h) reads 'establish and maintain in a licensed Malaysian financial institution one or more trust accounts, designated for the monies received from investors;' — verbatim, in SC-GL/6-2015(R13-2025) revised 6 January 2025. The SC list carries 'Updated list as of 26 June 2026' with the five named DAX. MOF carries the Section 4 crypto-tax sentence at 29 August 2025. The aFINity sentence is verbatim on sc.com.my/digital-assets. (Minor imprecision, not fatal: the Guidelines define 'licensed institution', not the phrase 'licensed Malaysian financial institution' used in 15.28(h) — the meaning is nonetheless plainly a licensed Malaysian bank.) MAPPING FAILS on the third conjunct. Anchor 4 requires 'Framework PLUS published tax treatment PLUS documented public-sector engagement.' The engagement conjunct rests SOLELY on aFINity: a September 2015 'Alliance of FinTech Community' launched to 'catalyse and nurture Fintech development in the Malaysian capital market' — fintech-wide, predating the 2019 digital-asset regime by four years, and appearing in the page's Contact Us footer. That is not documented public-sector engagement with digital assets. The 20 May 2026 guidelines revision appears only in the note with no fact URL, so it is not a verifiable fact under the contract's own rule, and it is in any event the regulator regulating — i.e. the framework limb restated. If a regulator merely operating its framework satisfied conjunct three, anchor 4 would collapse into anchor 3 and the conjunct would do no work; anchors are ceilings, not targets. This is the rubric's named upward-inference failure — 'an existing framework as public-sector adoption'. Anchor 3 ('regulated exchanges operate AND banking is obtainable') is fully and soundly carried: five DAX registered and operating, and the banking limb — which the rubric says a score cannot skip — is discharged by a binding instrument mandating trust accounts at licensed Malaysian banks rather than by inference. Score is 3. The note itself candidly predicted this exact knock-down. ORIGINAL SCORING NOTE, retained — its facts stand and only the mapping moved: FACTS: Licensed framework (SC Prescription Order 2019 + Guidelines on Recognized Markets, DAX chapter 15); five DAX registered and operating as of 26 June 2026; banking mandated at licensed Malaysian banks by para 15.28(h) — I read the instrument itself (SC-GL/6-2015(R13-2025), revised 6 January 2025; the SC's 20 May 2026 revision post-dates my copy and per the SC's own release only STRENGTHENS client asset safeguards, so 15.28(h) is not weakened); published tax treatment from MOF (29 Aug 2025); named SC programme aFINity. MAPPING: anchor 4 — all three conjuncts sourced, and anchor 3's banking limb (which the rubric says is the limb a score cannot skip) is carried by a binding instrument rather than inference. WHAT ARGUED HIGHER AND WHY IT FAILED: Anchor 5 requires active public-sector ADOPTION — a government programme, sandbox, or acceptance of digital assets. It does not reach: I found no Malaysian government acceptance of digital assets, and BNM's standing position is that crypto is not legal tender and not a regulated payment instrument. aFINity is engagement with a sector, not adoption of the asset. WHY IT MIGHT DROP TO 3: the third conjunct is the softest link — aFINity is a 2015 fintech-wide alliance, not a digital-asset-specific initiative, and a verifier reading "documented public-sector engagement" strictly (as requiring engagement beyond the regulator simply regulating) could knock this to 3. The framework and banking limbs are not in doubt. SOURCING GAP: LHDN's own guidelines PDF 404s at every URL the search index returns (hasil.gov.my media library appears reorganised), so the tax limb rests on MOF's primary statement rather than the LHDN instrument. TO MOVE UP: a named government digital-asset sandbox or public-sector acceptance/adoption. | seed | seed | f5ca1908-9d07-41a4-ba59-23c409e3bab0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 103 | 6c12d2a1-e457-4ab3-8cbf-a3d00832bfb7 | my | capital-controls | 1 | rubric 0-5 | https://www.investmalaysia.gov.my/media/rzzjreei/bnm-foreign-exchange-policy-notices.pdf | Bank Negara Malaysia (Foreign Exchange Policy Notices), mirrored on the Invest Malaysia government portal | 2022-06-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 1: Reporting or registration only; no approval needed; no documented delays THE FACTS: · BNM's FEP Notice 4, Part E, para 8 permits non-residents to repatriate income and divestment proceeds from Malaysia without any approval requirement — the only conditions are that repatriation be in foreign currency and that conversion go through the Notice 1 Part B channel. https://www.investmalaysia.gov.my/media/rzzjreei/bnm-foreign-exchange-policy-notices.pdf “A Non-Resident is allowed to repatriate from Malaysia, funds including any income earned or proceeds from divestment of Ringgit Asset, provided that— (a) the repatriation is made in Foreign Currency; and (b) the conversion of Ringgit into Foreign Currency is undertaken in accordance with Part B of Notice 1.” · The FEP regime is architecturally an approval regime — transactions in Schedule 14 of the FSA/IFSA are prohibited unless approved — but repatriation by non-residents is pre-approved by the Notices themselves, so no case-by-case approval arises for a node's normal operation. https://www.investmalaysia.gov.my/media/rzzjreei/bnm-foreign-exchange-policy-notices.pdf “A person shall obtain a written approval of the Bank to undertake or engage in any transaction listed in Schedule 14 of the FSA or IFSA that is not approved by the Bank under the FEP Notices.” · Non-residents may open and maintain External Accounts with Malaysian financial institutions. https://www.investmalaysia.gov.my/media/rzzjreei/bnm-foreign-exchange-policy-notices.pdf “A Non-Resident is allowed to open and maintain an External Account 4 with a Financial Institution in Malaysia.” FACTS: BNM's consolidated Foreign Exchange Policy Notices (issued 1 June 2022) allow non-residents to repatriate income and divestment proceeds freely, with no approval and no quantitative limit; conditions are form-and-channel only (must be in foreign currency, converted via a licensed onshore bank). I extracted and read the instrument text directly. MAPPING: anchor 1 — repatriation as of right, no approval, no documented delays found. SOURCE DISCIPLINE (rubric requires this disclosure): the rubric names IMF AREAER specifically. I could NOT use it — elibrary-areaer.imf.org redirects to an IMF bookstore auth gateway and is login-walled. I therefore reasoned from the central bank's own binding instrument rather than AREAER, which the rubric permits provided the note says so. This is that disclosure. A verifier with AREAER access should re-check. ADDITIONAL BLOCK: BNM's own copy of the Notices (bnm.gov.my/documents/20124/60360/Foreign+Exchange+Notices_Consolidated.pdf) returned HTTP 403, so I cite the identical document mirrored on the Invest Malaysia government portal and tier it OFFICIAL_SECONDARY accordingly — the document is BNM's, the host is not. WHAT ARGUED LOWER (better) AND WHY IT FAILED: anchor 0 ("No controls. Free repatriation") was arguable since repatriation needs no approval — but Malaysia is not a no-controls jurisdiction. There IS a control framework: a general Schedule 14 prohibition architecture with BNM approval as the gate, a foreign-currency-only requirement on repatriation (you cannot repatriate in ringgit), ringgit non-internationalisation, and reporting obligations to the Bank. That is "reporting or registration only" with conditions, i.e. anchor 1, not zero controls. WHAT ARGUED HIGHER: anchor 2 would need approval for large transfers — none is required for repatriation, so it does not reach. TO MOVE UP: reintroduction of case-by-case approval or conversion limits (Malaysia has form — see the 2016 NDF/export-conversion measures), which would push toward 2–4. CAVEAT: the copy I read is dated 2022-06-01; a later amendment may exist behind BNM's 403. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS HOLD. I extracted the BNM Foreign Exchange Policy Notices PDF directly. Notice 4 PART E para 8: 'A Non-Resident is allowed to repatriate from Malaysia, funds including any income earned or proceeds from divestment of Ringgit Asset, provided that— (a) the repatriation is made in Foreign Currency; and (b) the conversion of Ringgit into Foreign Currency is undertaken in accordance with Part B of Notice 1.' Preamble para 3: 'A person shall obtain a written approval of the Bank to undertake or engage in any transaction listed in Schedule 14 of the FSA or IFSA that is not approved by the Bank under the FEP Notices.' Notice 4 PART F para 9: 'A Non-Resident is allowed to open and maintain an External Account with a Financial Institution in Malaysia.' All verbatim; 'Issued on: 1 June 2022' as the note states. MAPPING: anchor 1 holds. Repatriation of income and divestment proceeds is pre-approved by the Notices themselves with no case-by-case approval and no quantitative limit, so anchor 2 ('approval required for large transfers') does not engage and anchors 3-5 are further off. Anchor 0 ('No controls') is correctly declined and the declining is the conservative direction on a lower-is-better axis: a real control architecture exists (Schedule 14 prohibition-unless-approved, foreign-currency-only repatriation, mandated conversion channel). SOURCE DISCIPLINE SATISFIED: the rubric names IMF AREAER and permits other reasoning provided the note says so — the note discloses the AREAER login-wall explicitly, and reasoning from the central bank's own binding instrument is stronger than the 'general posture' the rubric contemplates as the fallback. The BNM 403 and the Invest Malaysia mirror are disclosed and tiered OFFICIAL_SECONDARY, which is honest: the document is BNM's, the host is not. | seed | seed | cbb12e51-93ce-4ec5-8000-85a40f9d3810 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 104 | 9b0c8929-342a-420b-abce-bf07c05286da | my | policy-reversal-precedent | 4 | rubric 0-5 | https://www.kln.gov.my/web/chl_santiago/home/-/asset_publisher/9ECroxXNCJ6C/blog/temporary-suspension-of-malaysia-my-second-home-programme-mm2h- | Kementerian Pelancongan, Seni dan Budaya Malaysia (MOTAC), published via Ministry of Foreign Affairs / Embassy of Malaysia, Santiago | 2020-08-04 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A programme closed or suspended to new entrants with little notice THE FACTS: · THE INSTANCE: On 4 August 2020 the Government of Malaysia suspended the Malaysia My Second Home (MM2H) programme — a long-standing residence regime launched in 2002 — closing it to new entrants. https://www.kln.gov.my/web/chl_santiago/home/-/asset_publisher/9ECroxXNCJ6C/blog/temporary-suspension-of-malaysia-my-second-home-programme-mm2h- “The Government of Malaysia has decided to temporarily suspend the Malaysia My Second Home Program (MM2H) to allow Ministry of Tourism, Arts and Culture (MOTAC) and related agencies to comprehensively review and re-evaluate the MM2H program since its inception in 2002.” · The suspension caught applicants already in the queue — even fully submitted applications were halted, not processed under the terms they applied under. https://www.kln.gov.my/web/chl_santiago/home/-/asset_publisher/9ECroxXNCJ6C/blog/temporary-suspension-of-malaysia-my-second-home-programme-mm2h- “During the suspension, MOTAC is unable to process any new applications including applications that have already been submitted.” · Applicants in the queue were told to reapply later under whatever the new requirements turned out to be — no preservation of the terms they applied under. https://www.kln.gov.my/web/chl_santiago/home/-/asset_publisher/9ECroxXNCJ6C/blog/temporary-suspension-of-malaysia-my-second-home-programme-mm2h- “Therefore, foreigners who are still interested to participate in the MM2H Program are welcomed to reapply once the MM2H Program is reactivated but must abide to the current requirements.” · MM2H has been repeatedly re-termed since 2002 and now runs on a rebuilt four-category structure (Platinum, Gold, Silver, SEZ/SFZ), confirming the regime is one the state reopens on new terms rather than grandfathers. https://www.mm2h.gov.my/ “Having gone through several facelifts over the years since its launch in 2002, the current version of the MM2H Programme sees the introduction of 4 new categories—Platinum, Gold, Silver, as well as the Special Economic Zone/Special Financial Zone (SEZ/SFZ).” FACTS / THE INSTANCE CITED: MM2H, a residence regime running since 2002, was suspended to new entrants by MOTAC on 4 August 2020; applications already submitted were frozen unprocessed and those applicants were directed to reapply under new requirements once the programme reopened. It reopened in 2021 on materially tightened criteria. MAPPING: anchor 4 — "a programme closed or suspended to new entrants with little notice." This is a real, dated, cited instance in a regime old enough to supply evidence, so the rubric's "absence of history is UNRESOLVED, not a 0" trap does not apply. WHAT ARGUED HIGHER (worse) AND WHY IT FAILED: anchor 5 requires grants revoked from existing holders or a regime abolished without transition. Neither happened — existing MM2H pass holders were not revoked, and the regime was suspended and revived rather than abolished. On reopening, the new conditions were initially extended to existing holders with a one-year grace period (which would itself be anchor 3), but after stakeholder pressure existing holders were exempted from most of them, retaining only the higher processing fee and the 90-day annual residency requirement. WHAT ARGUED LOWER AND WHY IT FAILED: that walk-back to near-grandfathering is an anchor 2 story ("terms tightened prospectively, existing holders grandfathered") and is the strongest argument for a better score — but it does not erase the 2020 suspension, which is an independent and higher-scoring event. Anchor 4 is scored on the suspension, not on the grandfathering. TO MOVE DOWN: nothing available — the instance is on the record. TO MOVE UP to 5: revocation of existing passes or abolition of a regime without transition. SOURCING NOTE: mm2h.gov.my and nst.com.my return 403 to normal fetching and NST's body is JS-loaded; MOTAC's 4 August 2020 statement is reproduced verbatim and attributed on the Foreign Ministry's mission site, which is what I read and cite — hence OFFICIAL_SECONDARY (MOTAC issued it; MFA hosts it). CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS HOLD, all four — including the two the note disclosed as fetch failures, both of which I cleared and both of which resolve in the note's favour. kln.gov.my threw a TLS error to the fetcher, but the certificate is genuinely issued to 'O=Kementerian Luar Negeri Malaysia, CN=*.kln.gov.my' with merely an incomplete chain (server misconfiguration, not a suspect source); the page returns 200 and carries all three quotes verbatim: 'The Government of Malaysia has decided to temporarily suspend the Malaysia My Second Home Program (MM2H)...since its inception in 2002'; 'During the suspension, MOTAC is unable to process any new applications including applications that have already been submitted.'; 'foreigners who are still interested to participate in the MM2H Program are welcomed to reapply once the MM2H Program is reactivated but must abide to the current requirements.' Dated '4th August 2020', issued by MOTAC's Corporate Communications Unit — so the OFFICIAL_SECONDARY tiering (MOTAC issued, MFA hosts) is accurate. mm2h.gov.my 403s to the fetcher but returns 200 to a browser user-agent and carries the 'several facelifts over the years since its launch in 2002...Platinum, Gold, Silver, as well as the Special Economic Zone/Special Financial Zone (SEZ/SFZ)' sentence verbatim. MAPPING: anchor 4 ('A programme closed or suspended to new entrants with little notice') is squarely met by a dated, cited instance in a regime running since 2002, so the rubric's 'absence of history is UNRESOLVED, not a 0' trap does not apply. The COVID motive disclosed in the same statement does not carve out an exception the anchor does not contain, and the suspension went beyond a border closure: MOTAC froze already-submitted applications and directed applicants to reapply under whatever the new requirements became. Anchor 5 correctly declined — no revocation from existing holders, no abolition without transition. The near-grandfathering walk-back is an anchor 2 story about a different event and does not erase the suspension. | seed | seed | 178fd7d9-648e-49c6-b9a4-fe3b27189aef | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 105 | 5137b000-52c8-4f20-930e-00f9a9adb998 | th | visa-access | UNRESOLVED | NULL | https://image.mfa.go.th/mfa/0/n3gTFT2TOE/Visa_Requirements/Checklist_DTV.pdf | Ministry of Foreign Affairs, Kingdom of Thailand | 2024-07-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed {"US": [{"route": "Destination Thailand Visa (DTV) \u2014 Workcation category", "permits": ["RESIDE", "WORK_REMOTE"], "maxStayDays": 180}, {"route": "Destinatio and an independent agent re-fetched the source and rejected it. VERIFIER: Quote is verbatim (confirmed via pdftotext extraction of the MFA checklist). But the quote establishes almost none of the claimed matrix. The PDF is a document CHECKLIST only: it lists required documents for three DTV categories and is entirely SILENT on (a) maxStayDays — the number 180 appears nowhere; (b) the Non-Immigrant ED visa and its 90 days; (c) the LTR Work-from-Thailand Professional route and its 1825 days; (d) any RESIDE/WORK_REMOTE/STUDY permission grant; (e) any nationality distinction — it does not mention US, EU, IN, or CN, so it cannot support four identical per-nationality matrices. Worse, this checklist is Malaysia-specific (financial threshold given as '500,000 THB or (65,000 MYR)', requires 'a copy of the permanent resident in Malaysia (PR card)'), making it a poor basis for US/EU/IN/CN access. The category NAMES are supported; the permits, the day counts, and the ED and LTR routes are not. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: GATE FAILS — no single route permits RESIDE+STUDY+WORK_REMOTE for >=180 days. The near-miss is the DTV Workcation category: it permits RESIDE and WORK_REMOTE for 180 days but NOT STUDY. Reasoning, per route: (1) DTV Workcation — the MFA checklist heading (quoted) and every Royal Thai Embassy post define the category as 'Digital nomad/remote worker/foreign talent/freelancer'; remote work for a foreign employer is the express purpose, so WORK_REMOTE is permitted. NOTHING in the MFA checklist or in any official post page (Washington DC, Los Angeles, Yangon, Ho Chi Minh) says a DTV holder may enrol in an educational institution. Silence is not permission, so STUDY is NOT granted. The enabling instrument is titled 'Notification of the Ministry of Interior Re: Permission for Foreigners to Stay in the Kingdom for Tourism and Remote Work as a Special Case' (15 July 2024) — the title itself names tourism and remote work and not study. (2) DTV soft power category — permits attending activities, but the enumerated scope is 'Muaythai, Thai culinary training and medical treatment' (quoted). I map this to STUDY only in the loosest sense; it is a closed list of Thai soft-power activities and would NOT cover a foreign school's own curriculum. It carries no WORK_REMOTE, so it fails regardless. IMPORTANT: the DTV categories are separate qualification tracks — the applicant supplies category-specific documents (a foreign employment contract for Workcation vs. a letter of acceptance from the organising institute for soft power). Treating a Workcation DTV as also conferring study rights would be an inference, not a sourced permission. (3) Non-Immigrant ED — permits RESIDE+STUDY, but visa validity is 90 days and no official page grants remote work; fails on duration and on WORK_REMOTE. (4) LTR Work-from-Thailand Professional — permits RESIDE+WORK_REMOTE for 5 years (first grant), but BOI is silent on study, so no STUDY; also gated behind USD 80,000/yr income and a large-employer test. maxStayDays: DTV is 180 per entry; the +180 extension is EXCLUDED because it is discretionary ('may be granted', granted by the Immigration Bureau, '1-time ... not exceeding 180 days'), and the brief says not to count discretionary renewals. LTR 1825 = the 5-year first permission ('granted to stay in Thailand the first time for five years'); the further 5 years is excluded as discretionary. COHORT UNIFORMITY CAVEAT: I found no nationality restriction on the DTV in any official source; the MFA checklist is nationality-agnostic and segregates extra paperwork only under 'Additional Documents for SPECIFIC COUNTIRES'. I verified official posts covering US (Washington DC, Los Angeles) and Asia (Yangon, Ho Chi Minh, Taipei), and the New Delhi embassy confirms e-Visa availability in India from 1 Jan 2025. I did NOT find a China-post-specific DTV page; the CN row is asserted on the nationality-agnostic MFA checklist plus the universal e-Visa portal, which is weaker evidence than the US/EU/IN rows. Treat CN as the least-verified row. QUOTE THE RESEARCHER OFFERED: “A. Required Documents for a DTV: Workcation (Digital nomad/remote worker/foreign talent/freelancer) B. Required Documents for a DTV: Thai soft power related activities e.g. Muaythai, Thai culinary training and medical treatment C. Required Documents for a DTV: Spouse and children under 20 years old of DTV visa holders” | seed | seed | 6361a367-5bc2-4fba-969c-5f11bb820ebf | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 106 | 242b1dd6-816d-43f3-a798-59720eae77b7 | th | visa-access | UNRESOLVED | NULL | https://yangon.thaiembassy.org/en/publicservice/destination-thailand-visa-dtv?page=636b8dd1bd89964efa299793&menu=5d7d27c515e39c2e64001ff5 | Royal Thai Embassy, Yangon | 2024-07-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | SUPPORTING EVIDENCE for the DTV maxStayDays=180 figure in the matrix above — not an independent gate value. This official post states the per-entry stay is 180 days and that the extension is a single discretionary grant from the Immigration Bureau, after which the holder must leave and re-enter. This is why the matrix counts 180 and not 360. Value is UNRESOLVED because the row exists only to carry the duration quote; the gate value is the cohort matrix in the preceding observation. SOURCE QUOTE, verbatim from the page on the retrieval date: “Upon arrival, travelers with a DTV visa will be permitted to stay in Thailand for 180 days per entry.” | seed | seed | b130ddaa-007a-40df-b286-f52739d5f84c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 107 | d97b7fcc-6184-4d88-ac87-64cb993a5b6c | th | visa-access | {"route":"Destination Thailand Visa (DTV) — Workcation category","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":180,"passportSet":"*"} | NULL | https://yangon.thaiembassy.org/en/publicservice/destination-thailand-visa-dtv?page=636b8dd1bd89964efa299793&menu=5d7d27c515e39c2e64001ff5 | Royal Thai Embassy, Yangon (Ministry of Foreign Affairs post) | 2024-07-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the page on the retrieval date: "This type of visa is issued to applicants who wish to enter Thailand for workcation (digital nomad/remote worker/foreign talent/ freelancer) purposes" — WORK_REMOTE is the route's express purpose, not an inference. "Validity of Visa: 5 years Multiple Entries". "Upon arrival, travelers with a DTV visa will be permitted to stay in Thailand for 180 days per entry." — maxStayDays 180 is the sourced PER-ENTRY stay, not the 5-year visa validity. EXTENSION NOT COUNTED, with the rule quoted per the Vietnam-pass correction: "Those who wish to have an Extension of Stay through the Thai Immigration Bureau located at Government Center B, Chaengwattana Soi 7, Laksi, Bangkok 10210, Tel 0-2141-9889 (or at http://www.immigration.go.th). 1-time for a period not exceeding 180 days per entry. After the maximum stay of 180 days + 180 days needs to depart the country and re-enter Thailand with the same DTV visa within the visa validity date." — the page routes the applicant to the Immigration Bureau and states a 1-time cap; it states NO as-of-right entitlement, and I found no published Immigration Bureau order granting it as of right, so the extension is treated as discretionary and 180, not 360, is the honest figure. WHAT IT DOES NOT PERMIT: STUDY — nothing on this or any official DTV page grants enrolment in an educational institution to a Workcation holder, and the enabling instrument is titled (per secondary attestations; Gazette text unread, see notes) for tourism and remote work; silence is not permission. Grant discretion: "The consular officer reserves the right to request additional documents and do an interview as deemed necessary." PASSPORT-SET BASIS FOR "*": the WEAKER form, stated explicitly per PASSPORT.md §1 — this page states document and financial conditions (500,000 THB) with no nationality condition, and I did NOT find an affirmative statement that the DTV is open to all nationalities. Corroboration that the route is served to the scenario's cohorts: US posts (Washington DC, Los Angeles), an EU post (Helsinki), the New Delhi embassy (announced the DTV among measures effective 15 July 2024), and a China post (Chengdu Consulate-General FAQ, updated 4 Nov 2025) all publish the route — see researcher-notes.md. TIER GAP DISCLOSED: an embassy post restating the Ministry of Interior notification is OFFICIAL_SECONDARY; the enabling instrument (Notification of the Ministry of Interior re: permission to stay for tourism and remote work as a special case, 15 July 2024) was NOT retrieved — ratchakitcha.soc.go.th is reachable but serves PDF downloads this fetcher cannot render. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, yangon.thaiembassy.org, page updated 18 Jul 2024. All claim quotes present verbatim. 180 is the per-entry stay, distinct from the 5-year validity. The extension is routed to the Immigration Bureau with no as-of-right language, so 180 (not 360) is correct. No nationality condition appears anywhere on the page — the weaker '*' basis is accurately characterized. WORK_REMOTE is the route's express purpose; STUDY is nowhere granted. | seed | seed | c862678d-5f08-42f3-92f8-76e33a082a7d | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 108 | 760deed8-788a-4473-92c6-139e1fc4abe3 | th | visa-access | {"route":"Destination Thailand Visa (DTV) — Workcation category","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":180,"passportSet":"*"} | NULL | https://helsinki.thaiembassy.org/en/publicservice/destination-thailand-visa-dtv?menu=5d80876d15e39c3354007bb1&page=5f49f4a199a85e260f4278de | Royal Thai Embassy, Helsinki (Ministry of Foreign Affairs post) | 2024-07-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Second observation of the same route from an independent post (an EU-country post, page updated 15 Oct 2025), filed because it carries the terms in one block and the extension wording in its weakest honest form. SOURCE QUOTES, verbatim: "VISA Fee : 350 Euro / VISA validity : 5 years Multiple entries / Period of Stay : 180 days per entry / Extension of Stay : for another 180 days is possible at the Thai Immigration Bureau. After the maximum stay of 180+ 180 days, DTV holders are required to depart from Thailand and re-enter Thailand again within the visa validity." — "is possible" is availability, not entitlement; the extension is NOT counted into maxStayDays. Purpose list, verbatim: "Workcation (Digital nomad/remote worker/foreign talent/freelancer) / Thai soft power related activities e.g. Muaythai, Thai culinary training and medical treatment / Spouse and children under 20 years old of DTV visa holders". Workcation category document, verbatim: "Employment contract or employment certificate in their country or professional portfolio showcasing digital nomad, remote worker, foreign talent or freelancer status" — the employer is in the applicant's country; WORK_REMOTE for a foreign employer is the sourced grant. Discretion, verbatim: "All applications are subject to approval and the Royal Thai Embassy may request for an interview or additional documents. Submission of a visa application does not necessarily mean that a visa will be granted." NOT PERMITTED: STUDY — not granted anywhere on the page. PASSPORT-SET BASIS FOR "*": weaker form — no nationality condition stated; the page's only nationality-shaped line is a residence-jurisdiction rule ("For non-Finnish and non-Estonian citizens, please provide ... a valid residence permit card"), which conditions WHERE you apply, not WHICH passport may hold the route. No affirmative universality statement found; said explicitly per PASSPORT.md §1. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, helsinki.thaiembassy.org, page updated 15 Oct 2025. All claim quotes present verbatim, including the Workcation document line ('Employment contract or employment certificate in their country or professional portfolio...'), the discretion clause, and the purpose list. 'is possible' is availability language, not entitlement — excluding the +180 from maxStayDays is correct. The only nationality-shaped line is the non-Finnish/non-Estonian residence-permit requirement, which conditions where you apply, not which passport qualifies — exactly as the claim characterizes it. | seed | seed | 2192290f-e4d5-47f5-8bb2-c78cafb72c44 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 109 | b5a11202-db2c-40b8-a2a8-44509db39d08 | th | visa-access | {"route":"Destination Thailand Visa (DTV) — Workcation category","permits":["WORK_REMOTE"],"maxStayDays":null,"passportSet":"*"} | NULL | https://washingtondc.thaiembassy.org/en/page/dtv-visa | Royal Thai Embassy, Washington, D.C. (Ministry of Foreign Affairs post) | 2024-07-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SUPPORTING OBSERVATION, deliberately partial: this page does NOT state the 180-day per-entry stay, so maxStayDays is null here and the day count rests on the Yangon and Helsinki observations — I will not attach a number to a page that does not carry it (the exact defect the D-032 verifier rejected in the prior Thailand matrix). WHY IT IS FILED ANYWAY: currency. The page's own header reads "วันที่ปรับปรุงข้อมูล 19 Mar 2026" (last updated 19 Mar 2026), which evidences that the DTV remained a live, published route AFTER the 19 May 2026-announced rollback cycle began and after the 2025 e-Visa migration — the freshest official DTV statement found, and it is a US post, which also corroborates availability to the US cohort. SOURCE QUOTES, verbatim: "Validity of visa: 5 years"; "Purpose of Visit: Workcation (Digital nomad/remote worker/foreign talent/freelancer) / Thai soft power related activities e.g. Muaythai, Thai culinary training and medical treatment / Spouse and children under 20 years old of DTV visa holders"; financial test "ending balance of no less than 500,000 THB or ($16,000 USD)"; "*For an extension of stay, please contact the Immigration Bureau directly once you are in Thailand*" — again no as-of-right extension rule. The "For non-US citizens" additional-document list shows the post serves non-US passports resident in the US and, by its own structure, that US citizens apply on the base document list — affirmative evidence the route is open to the US passport, though the route itself is stated without any nationality condition (weaker "*" basis, as in the sibling observations). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, washingtondc.thaiembassy.org/en/page/dtv-visa. The absence claimed is real: I read the full rendered page and it states NO per-entry stay figure anywhere — maxStayDays null is correct and attaching 180 to this URL would have been a defect. The 19 Mar 2026 update date is on the page verbatim. The 'For non-US citizens' additional-documents block exists and structurally implies US citizens apply on the base list. Deliberately-partial permits [WORK_REMOTE] is defensible restraint: with no stay period stated on this page, not recording RESIDE here is the conservative reading, and the sibling claims carry it. | seed | seed | ab2ecceb-e31e-4843-8ba7-0ff58bfa975e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 110 | b70143b4-0d49-49b1-8358-1a70db09641a | th | visa-access | {"route":"Destination Thailand Visa (DTV) — Thai soft power category","permits":["RESIDE"],"maxStayDays":180,"passportSet":"*"} | NULL | https://helsinki.thaiembassy.org/en/publicservice/destination-thailand-visa-dtv?menu=5d80876d15e39c3354007bb1&page=5f49f4a199a85e260f4278de | Royal Thai Embassy, Helsinki (Ministry of Foreign Affairs post) | 2024-07-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | The DTV's second qualification category, filed as its own route because the categories carry separate document tracks and are applied for as distinct purposes. SOURCE QUOTES, verbatim: category name "Thai soft power related activities e.g. Muaythai, Thai culinary training and medical treatment"; qualifying document "Letter of acceptance from institute or company organizing activities, letter of appointment from hospital/medical center"; stay "Period of Stay : 180 days per entry" (the page states one stay rule for the DTV as a whole). WHY permits IS [RESIDE] AND NOT [RESIDE, STUDY]: the sourced scope is attendance at Thai soft-power activities run by an organizing institute (Muaythai, Thai culinary training, medical treatment — an "e.g." list, so open-textured, but every named example is a Thai cultural/wellness activity). Mapping that to the scenario's STUDY — enrolment in a foreign school's own curriculum — would be an inference the source does not support, so STUDY is NOT recorded; this is stated as a deliberate under-claim, and the STUDY question is recorded as unknown rather than no. NOT PERMITTED: WORK_REMOTE — the workcation documents (foreign employment contract) belong to the other category, and no official page states that a soft-power DTV holder may work remotely, nor that one holder may exercise both categories' purposes on one visa; whether the category is fixed at issuance is UNKNOWN — no official source found either way, and I did not stretch (the deciding question the tasking flagged). PASSPORT-SET BASIS FOR "*": weaker form — no nationality condition stated; no affirmative universality statement found. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, helsinki.thaiembassy.org (same page as claim 1). The page states one stay rule (180/entry) for the DTV as a whole, covering this category. Recording [RESIDE] without STUDY is a defensible deliberate under-claim: the sourced scope is attendance at organizer-run Thai cultural/wellness activities, and the scenario's STUDY (a foreign school's own curriculum) is not what the page grants; the researcher recorded the STUDY question as unknown rather than no, which is the honest shape. Not recording WORK_REMOTE is correct — the foreign-employment documents belong to the Workcation track only. | seed | seed | 8cdb355e-1543-43ef-b72d-33bb2bfc974f | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 111 | 91129760-d10d-4a42-ac18-1c5e100aaa9d | th | visa-access | {"route":"Long-Term Resident Visa (LTR) — Work-from-Thailand Professionals category","permits":["RESIDE","WORK_REMOTE"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://ltr.boi.go.th/ | Thailand Board of Investment (BOI) — LTR Visa Unit | 2022-09-01 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the BOI's own LTR site on the retrieval date: "There are four types of LTR visas designed to facilitate different types of individuals: Wealthy Global Citizen, Wealthy Pensioners, Work-from-Thailand Professionals, and Highly skilled Professionals." Stay: "10* years renewable visa Permission will be granted to stay in Thailand the first time for five years which can be extended for 5 more years if qualifications are met." — maxStayDays 1825 is the FIRST permission (5 years); the further 5 years is conditional ("if qualifications are met") and adjudicated at renewal, so it is NOT counted, per the same rule that excluded the DTV extension. Note the 1825 is a permission-to-stay term, not a per-entry counter — the visa carries "Multiple re-entry permit" and "90-day report extended to 1-year report", so 1825 ≥ 180 is the honest comparison for requiredStayDays. WORK_REMOTE is express, quoted: "Work permit will not be granted for Work-From-Thailand Professionals holders since foreigners under this LTR Visa category are applying for the purpose of working for foreign employer abroad remotely from Thailand. Therefore, the person do not have Thai employers according to the digital work permit purpose." NOT PERMITTED: STUDY — the site is silent on enrolment; silence is not permission. ELIGIBILITY GATES (why this is not a general-cohort route): personal income of USD 80,000/yr in the past two years (USD 40,000 with a master's degree/IP/Series A), an overseas employer meeting a corporate-scale test (lowered to USD 50M revenue over three years by the Cabinet update announced 15 Jan 2025 — see the companion OSOS observation), and health insurance/deposit conditions. effectiveAt 2022-09-01 rests on the Royal Thai Embassy Washington D.C. LTR page (https://washingtondc.thaiembassy.org/en/page/ltr), verbatim: "By the order of the Ministry of Interior of the Kingdom of Thailand dated 2 June B.E. 2565 (2022). Thailand has launched a new Long-Term Resident Visa (LTR), which will come into effect on 1 September 2022 onwards." — disclosed because it is a second document; the MoI order itself was not retrieved. PASSPORT-SET BASIS FOR "*": weaker form — the criteria are income/employer/insurance tests with no nationality condition stated anywhere on the page; no affirmative universality statement found. The companion OSOS observation adds affirmative multi-nationality GRANT data (Europe, US, China, India all among recipients). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, ltr.boi.go.th. All claim quotes verbatim: the four-types sentence, the five-year first permission, the WFTP remote-work rationale, 'Multiple re-entry permit' and '90-day report extended to 1-year report'. 1825 = 5 years first permission is correct and correctly excludes the conditional second 5 years ('if qualifications are met'). WORK_REMOTE is express in the page's own words. The supporting effectiveAt document was independently re-fetched: washingtondc.thaiembassy.org/en/page/ltr states verbatim 'By the order of the Ministry of Interior of the Kingdom of Thailand dated 2 June B.E. 2565 (2022). Thailand has launched a new Long-Term Resident Visa (LTR), which will come into effect on 1 September 2022 onwards.' — and also corroborates the WFTP USD 80,000/40,000 income criteria the note cites. No nationality condition on either page. | seed | seed | 062c36c2-783b-428c-8176-dfeace9588e9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 112 | 22a0b645-41a5-4c34-8d14-8e9cb4e00979 | th | visa-access | {"route":"Long-Term Resident Visa (LTR) — Work-from-Thailand Professionals category","permits":["RESIDE"],"maxStayDays":null,"passportSet":"*"} | NULL | https://osos.boi.go.th/EN/news/2158/Thailand-Approves-Update-of-Long-Term-Resident-LTR-Visa%E2%80%99s/ | Thailand Board of Investment — One Start One Stop Investment Center (OSOS) | 2025-01-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SUPPORTING OBSERVATION, deliberately partial: maxStayDays is null because this page states "a 10-year residence" as the program headline and no per-grant day figure; the 5-year first-permission figure rests on the ltr.boi.go.th observation. WHY IT IS FILED: it carries the strongest nationality evidence found for the LTR — affirmative GRANT data across the scenario's cohorts, verbatim: "As of the end of 2024, the BOI has already granted LTR visas to more than 6,000 applicants. Europe leads the list with 2,500 recipients, followed by the United States (1,080), Japan (610), China (340), and India (280)." — this is evidence of grants to European, US, Chinese and Indian nationals, which is stronger than mere absence of a nationality condition, though still NOT a statement of universal eligibility (said explicitly per PASSPORT.md §1; "Europe" here is the BOI's own word, not a bloc I coined, and it is regional grant statistics, not a rule keyed to EU/EEA/CH). Also carries the 15 Jan 2025 Cabinet-approved criteria changes for Work-from-Thailand Professionals, verbatim: "The corporate revenue requirement for overseas-based employers sponsoring an applicant has been lowered to a minimum of USD 50 million over the last three years, down from USD 150 million." and "The requirement for a minimum of five years of work experience in a relevant field for applicants to the Highly Skilled Professionals and Work-from-Thailand Professionals categories has been removed", with the caveat quoted: "All changes will take effect following the official BOI announcement on the updated LTR criteria and the Ministry of Interior's announcement on the expansion of dependents' rights." — I did not verify that those implementing announcements have since issued; the criteria figures should be treated as the announced direction, not confirmed in-force values. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, osos.boi.go.th news 2158, page dated 15 January 2025. All quotes verbatim: '10-year residence' headline, the grant statistics, the USD 50M employer-revenue reduction ('down from USD 150 million'), the removal of the five-year experience requirement, and the implementing-announcement caveat. maxStayDays null is correct — the page states no per-grant day figure. 'Europe' is the BOI's own word, quoted not coined. The claim's caution that the Jan 2025 criteria are announced direction, not confirmed in-force, matches the page's own conditional wording. | seed | seed | b017acad-9e47-474b-8914-bf0bfe72c156 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 113 | 66185918-37be-4463-bbc2-7fa27c39427c | th | visa-access | {"route":"Non-Immigrant Visa ED (study below bachelor's degree / seminar / training / curricular internship)","permits":["RESIDE","STUDY"],"maxStayDays":90,"passportSet":"*"} | NULL | https://hochiminh.thaiembassy.org/en/publicservice/non-immigrant-ed-studying-attending-seminar-traini?cate=5d809e6e15e39c2fe80078a2 | Royal Thai Consulate-General in Ho Chi Minh City (Ministry of Foreign Affairs post) | 2025-04-01 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the page (updated 1 Apr 2025): "This type of visa is issued to applicants who wish to study (below bachelor's degree), attend seminar, training session or curricular internship in Thailand." Stay: "Holder of this type of visa is entitled to stay in Thailand for a maximum period of 90 days. He or she may apply for an extension of stay at the Office of the Immigration Bureau at Government Center B, Chaengwattana Soi 7, Laksi, Bangkok 10210 ... and may be granted such extension for a period of one year from the date of first entry into Thailand." — maxStayDays 90; the one-year extension is "may be granted", i.e. DISCRETIONARY on the page's own words, so it is not counted (the London post states the same in terms: "The extension of stay is at the discretion of the immigration officer." — see the companion London observation). FAILS the scenario on duration (90 < 180) before activities are even weighed. WHAT IT PERMITS: STUDY at a Thai-approved establishment plus lawful stay (RESIDE for the permitted period). WHAT IT DOES NOT PERMIT: WORK_REMOTE or any work — quoted, verbatim: "An alien who receives a Non-Immigrant visa can work in Thailand once he or she is being granted a work permit. An alien in violation of the Immigration Act B.E. 2522 (1979) concerning taking up employment without work permit or the Royal Decree B.E.2522 (1979) concerning holding employment in certain restricted occupations and professions shall be prosecuted and imprisoned or fined, or shall face both penalties." — no work permit attaches to ED, and remote work for a foreign employer is nowhere granted. Grant discretion: "The Royal Thai Consulate-General in Ho Chi Minh reserves the rights to request additional documents, interviews, or to disapprove as deemed necessary." effectiveAt DISCLOSURE: 2025-04-01 is the page's own last-update date; the ED category's original enabling instrument date was not sourced in this pass and I would not invent one. PASSPORT-SET BASIS FOR "*": weaker form — no nationality condition stated; no affirmative universality statement found. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, hochiminh.thaiembassy.org, page updated 1 Apr 2025. All quotes verbatim: the purpose sentence ('study (below bachelor's degree), attend seminar, training session or curricular internship'), the 90-day entitlement, and the Immigration Act B.E. 2522 work-permit warning. maxStayDays 90 matches; the one-year extension is 'may be granted' — discretionary on the page's own words, correctly not counted. No work of any kind is granted without a separate work permit, so omitting WORK_REMOTE is correct. No nationality condition stated. | seed | seed | fc4b394a-f6ce-4175-9492-f6502016c0e9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 114 | 0277f17f-81cb-4f9b-a640-d53fc64a5f0c | th | visa-access | {"route":"Non-Immigrant Visa ED (study below bachelor's degree / seminar / training / curricular internship)","permits":["RESIDE","STUDY"],"maxStayDays":90,"passportSet":"*"} | NULL | https://london.thaiembassy.org/en/page/education-visa | Royal Thai Embassy, London (Ministry of Foreign Affairs post) | 2024-07-17 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Second observation of the ED route from an independent post, filed for two clauses the Ho Chi Minh page does not carry. (1) THE NON-FORMAL-SCHOOL EXCLUSION, verbatim: "This category of visa is issued to applicants who are students of educational establishments in Thailand approved by Thai authorities except for students of non-formal schools in accordance with laws on private schools or of any other similar private educational establishments." — on this post's stated rule, ED visas are NOT issued to students of non-formal schools under the Private School Act, which is precisely the statutory category a Network School node would most plausibly occupy (see the operating-licence observation in th.json: PSA s.120 non-formal school licensing). If that exclusion is the national rule and not one post's wording, members could not use ED to study at the node even for 90 days — I did not find the rule stated by a second post or the MFA centrally, so its generality is UNKNOWN and recorded, not asserted. (2) EXTENSION DISCRETION IN TERMS, verbatim: "The extension of stay is at the discretion of the immigration officer." — the published ground for refusing to count the 1-year ED extension into maxStayDays. Stay, verbatim: "The ED Visa is valid for 90 days from the date of issuance. Travellers must enter Thailand within the said 90-day period. They will be allowed to stay in Thailand for a period of up to 90 days, counting from the date of entry." — maxStayDays 90, single entry. effectiveAt DISCLOSURE: 2024-07-17 is the page's own last-update date; the category's enabling instrument date was not sourced. PASSPORT-SET BASIS FOR "*": weaker form — no nationality condition stated; no affirmative universality statement found. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, london.thaiembassy.org/en/page/education-visa, page updated 17 Jul 2024 (matching the claim's effectiveAt). All three load-bearing quotes verbatim: the non-formal-school exclusion, the 90-day single-entry stay, and the express discretion sentence. The claim correctly records the exclusion's generality as UNKNOWN rather than asserting it as the national rule — it appears on this post's page only among the pages fetched. maxStayDays 90 confirmed as a per-entry stay distinct from the 90-day entry validity window. | seed | seed | 5ed68f12-04ef-4d37-98e0-a8f6637a9190 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 115 | 5d0b7b34-de2a-4d67-8e5c-19424f2879da | th | visa-access | {"route":"Non-Immigrant Visa ED Plus (study at bachelor's degree level or above at a Thai university/institution)","permits":["RESIDE","STUDY"],"maxStayDays":null,"passportSet":"*"} | NULL | https://hochiminh.thaiembassy.org/en/publicservice/non-immigrant-ed-studying-attending-seminar-traini?cate=5d809e6e15e39c2fe80078a2 | Royal Thai Consulate-General in Ho Chi Minh City (Ministry of Foreign Affairs post) | 2024-07-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | maxStayDays IS NULL BECAUSE THE SOURCE STATES NO DAY COUNT and I will not invent one: the page's period-of-stay section says, verbatim, "During the course of study, foreign students in bachelor degree level or higher can depart and re-enter Thailand without the need for a re-entry permit." and "Universities/Institutions will submit the visa extension application on behalf of the students." — a study-duration-shaped stay with no published per-entry number on this page. Eligibility, verbatim: "This type of visa is applied for applicants who are international students currently studying at universities/institutions in Thailand at a bachelor's degree level or above." — the route is bound to enrolment at a THAI university/institution at degree level; it could not cover study at a foreign school's own non-degree curriculum, so even at unlimited duration it would not answer the scenario's STUDY as Network School delivers it unless the node were itself a Thai degree institution (which the operating-licence research shows it could not straightforwardly be). Post-graduation, verbatim: "Foreign students are allowed to extend the stay in Thailand for 1 year after graduating to seek employment, travel or participate in other activities". NOT PERMITTED: work — the same Immigration Act B.E. 2522 work-permit warning is appended to this section verbatim as to ED. effectiveAt 2024-07-15 rests on the Royal Thai Embassy New Delhi announcement (https://newdelhi.thaiembassy.org/en/content/new-thailand-s-visa-measures?cate=614489badd57933bb40edbf2), verbatim: "1. Visa Exemption (60 days) for 93 countries/territories 2. Visa on arrival (VoA) for 31 countries/territories 3. The Destination Thailand Visa (DTV) 4. The Non-Immigrant (Education - ED Plus) ... Note: Effective from 15 July 2024 onwards" — disclosed as a second document. PASSPORT-SET BASIS FOR "*": weaker form — no nationality condition stated; no affirmative universality statement found. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, hochiminh.thaiembassy.org (same page as claim 6, ED Plus section). The null day count is a REAL absence, not a missed line: I read the full ED Plus section and its Period of Stay states no per-entry day figure; the only numbers are 'Validity of Visa: 3 months' (visa validity, not stay) and the 1-year post-graduation extension (correctly not counted into maxStayDays). The route is bound to Thai degree-level enrolment as claimed. The supporting effectiveAt document was independently re-fetched: newdelhi.thaiembassy.org states verbatim '4. The Non-Immigrant (Education - ED Plus) ... Note: Effective from 15 July 2024 onwards'. | seed | seed | eed222f8-2d99-46f2-8f1b-ca62f2b1a627 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 116 | 4e641131-467b-482a-bb78-7e4e233cf362 | th | operating-licence | false | NULL | https://natlex.ilo.org/dyn/natlex2/natlex2/files/download/80405/THA80405%20Eng.pdf | Private School Act B.E. 2550 (2007), Kingdom of Thailand (official English translation, hosted by ILO NATLEX) | 2007-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | DOCUMENT_EXTRACTION | NULL | A foreign entity may NOT hold a private school licence in Thailand as a matter of statute. Chain of reasoning, all from the quoted Act text: Section 120 — establishing a Non-formal School requires a licence from the Licensor. Section 21(1) — a natural-person applicant for a school establishment licence must 'Be of Thai Nationality'. Section 22(2) — a juristic-person applicant (company limited / public company / partnership) must have not less than HALF its shares or capital held by shareholders of Thai nationality. Critically, Section 127(1) applies Sections 21 and 22 to NON-FORMAL Schools mutatis mutandis — so the Thai-majority requirement is NOT escapable by positioning Network School as a non-formal/flexible institution rather than a degree-granting one. Section 37 additionally requires the school Director to be of Thai nationality. TWO OPEN QUESTIONS I could not close and which a lawyer must resolve before this is treated as final: (1) BOI promotion is claimed by several law firms to permit 100% foreign ownership in education, but I found no primary instrument, and it is not obvious that a BOI promotion under the Foreign Business Act overrides the PSA's own licensee-nationality test, which is a separate statutory requirement; (2) whether Network School is a 'School' within the meaning of the Act at all — Section 120 para 2 leaves the category and nature of a Non-formal School to be prescribed by the Commission and published in the Government Gazette, which I did not retrieve. If Network School delivers no prescribed curriculum it might fall outside the Act, in which case it would need no licence but would also have no licensed status. Recorded false, not UNRESOLVED, because the licensing regime was actually located and it is explicit; but the value is the statute's answer, not a legal opinion. Effective date is the Act's year B.E. 2550 = 2007; exact promulgation date not extracted, and the text used is 'up to amended No. 2' (B.E. 2554 / 2011). SOURCE QUOTE, verbatim from the page on the retrieval date: “Section 21. An applicant for a license to establish a Formal School being a natural person shall have the qualifications and have no prohibitions as follows; (1) Be of Thai Nationality; [...] Section 22. An applicant for a license to establish a Formal School which is a juristic person shall have the qualifications and have no prohibitions as follows; (2) A juristic person being a public company limited, company limited or partnership shall have the number of shares or capital belonging to the shareholders or partners of Thai Nationality of not less than a half of the total number of shares or capital. [...] Section 120. The establishment of a Non-formal School shall require a license from the Licensor. [...] Section 127. The following provisions including related penal provisions shall apply to the Non-formal Schools mutatis mutandis. (1) Section 21, Section 22, Section 28, Section 29, Section 33, paragraph one and paragraph three of Section 43 and Section 48 (3), (4) and (5);” | seed | seed | 48c7ac10-e790-401d-ba17-09ace16f2a13 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 117 | 22075e39-6d12-4da5-89ed-e31917ff8ecf | th | foreign-title-statute | UNRESOLVED | NULL | https://faolex.fao.org/docs/pdf/tha33176.pdf | Land Code Promulgating Act, B.E. 2497 (1954), Kingdom of Thailand (English translation, hosted by FAO FAOLEX) | 1954-01-01 | 2026-07-17T00:00:00.000Z | MULTILATERAL | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | Partially sourced, deliberately not resolved. The gate asks whether a foreign entity may HOLD OR LEASE property as a matter of national statute — an OR. I sourced the HOLD limb and could not source the LEASE limb, so the gate stays UNRESOLVED. HOLD limb (sourced, quoted): Land Code Section 86 permits aliens to acquire land only 'by virtue of the provisions of a treaty giving the right to own immovable properties'. Secondary sources state no such treaty is currently in force, which would make foreign freehold effectively unavailable — I did not verify that treaty claim against a primary source and do not assert it. Section 86 para 2 does contemplate aliens acquiring land for 'commerce' with the permission of the Minister under Ministerial Regulations, but it is expressly subject to Section 84 and to para 1's treaty condition; Section 87 caps commercial acquisition at 1 rai. Section 96 bis adds a narrow residential exemption. This is lawyer territory and I will not score it. LEASE limb (NOT sourced): leasehold is governed by the Civil and Commercial Code (reportedly ss.537-571, 30-year maximum), which is a different statute — I grepped the full Land Code text and it does not govern foreign leasing. Every source I found for the 30-year lease cap was a law firm or property agency, which is not good enough for a gate. To close this, fetch the Civil and Commercial Code Book III Title IV (Hire of Property) from a primary source and confirm whether a foreign juristic person may be a lessee and on what term. SOURCE QUOTE, verbatim from the page on the retrieval date: “Section 86 Aliens may acquire land by virtue of the provisions of a treaty giving the right to own immovable properties and subject to the provisions of this Code. Subject the Section 84 the aforesaid aliens may acquire land for residence, commerce, industry, agriculture, burial, public charity or religion under the conditions and procedures prescribed in Ministerial Regulations and with the permission of the Minister.” | seed | seed | c5959a2d-66e5-41f3-8869-01fbace3a8a0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 118 | cb93d59b-37d1-46b1-8097-d1bcb3e78a4f | kz-almaty-city | flood-sea-level | 5 | rubric 0-5 | https://cyberleninka.ru/article/n/selevye-potoki-v-basseynah-rek-malaya-i-bolshaya-almatinka | Medeu A.R. (Director, Institute of Geography, MON RK), Blagoveshchensky V.P., Ranova S.U. — debris-flow regime of the Malaya and Bolshaya Almatinka basins (2019) | 2019-01-01 | 2026-07-22T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | ANALYST_ENTRY | NULL | OWNER SCOPE RULING D-052 IS WHAT MAKES THIS SCOREABLE: rubric.md §8 covers debris-flow inundation, not only coastal and fluvial flooding. Without that ruling this criterion stays NEEDS_EVIDENCE, which is exactly where the previous verifier correctly parked it — the facts were never in doubt, the anchor set's scope was. SCALE (rubric.md §8, RISK axis, LOWER is better): 5 is the most exposed. ANCHOR MAPPING -> 5, on anchor 5's THIRD disjunct, 'documented history of inundation at this site', whose sourcing column names 'incident reports'. Almaty's downtown occupies the deposition fan of the Malaya Almatinka, and the inundation history is documented and catastrophic: the 1921 mudflow killed 500 of a population of 45,000 and destroyed much of the city; five catastrophic debris flows occurred in the two basins during the 20th century; the 1973 event was caught by the Medeu dam. THE RETURN PERIODS, verbatim from the 2019 paper and re-fetched by the adversarial verifier (19,076 characters, every quoted string literally present): down the main channel of the Kishi (Malaya) Almatinka, large and very large mudflows THREATENING THE TERRITORY OF THE CITY occurred 11 times in 300 years — a mean recurrence of about 30 years, i.e. a 3.3% annual exceedance probability. Ulken (Bolshaya) Almatinka: 13 in 214 years. The paper's own zonation scheme is expressed in exactly the terms §8 asks for — frequent >2%/yr, rare 1-2%, very rare <1% — mapped as 'границы распространения селей различных объемов и повторяемости'. Tributary recurrences are given too (Gorelnik 10%, Bedelbay/Batareyka 12%, Butakovka ~1-in-100). WHY NOT 4: the verifier established that anchor 4 is STRUCTURALLY UNREACHABLE here — its first disjunct is a conjunction requiring coastal exposure below 5 m, and Almaty is 2,000 km from the nearest sea at 700-900 m. WHY THE MEDEU DAM DOES NOT DISCOUNT THIS, and this is the uncomfortable part of the ruling: the defences are real, are described in the source as 'надежной защитой' with a multiple safety margin, and were tested successfully in 1973. But §8 grants a defences discount ONLY on anchor 3's coastal disjunct, so under the contract as written the dam is RUBRIC-INVISIBLE. That is a contract shape, not a judgment about Almaty's engineering, and it is recorded here so nobody reads this 5 as a claim that the city is undefended. LIVE RESIDUAL SOURCES are named by the same source: lakes No. 6 and No. 13-bis, the latter capable of a 2 million m3 flow. WHAT THE FLOOD-ONLY INSTRUMENTS SAY, recorded because it is the counter-case: ThinkHazard! returns river flood 'Very low' AND urban flood 'Very low' for Almaty (one class BELOW Tashkent's 'Low', which scored 1), with landslide 'Medium'. Under a flood-only reading of §8 this criterion would be 0 or 1. The whole distance between 0 and 5 here is definitional, which is why the verifier flagged — and the owner has recorded as open — that debris flow may deserve its own criterion rather than being litigated into a sea-level anchor set. LICENCE: CC BY-ND — quotable, not re-derivable. VENUE, disclosed rather than hidden: a Russian pay-to-publish monthly; the warrant is the authors, the Director and staff of the Institute of Geography MON RK, not the venue. CORROBORATED independently by Kazhydromet's own journal (Abdullayeva & Mussina 2023, CC BY-NC), which publishes a 1:25,000 mudflow hazard and vulnerability map for the same two basins and confirms the 2019 paper's issue and pages. STILL UNRETRIEVED: the Kazselezashchita / Kazhydromet fan-zonation MAP FOOTPRINT itself — gov.kz timed out at 45 s and Kazselezashchita has no independently resolving public site. That map is what a property-level call on a specific address would need. PROPERTY-LEVEL FLAG, carried forward verbatim from the first pass: a node sited ON the fan in the upper southern districts (Medeu, the Gorniy Gigant / Koktobe slopes) is materially more exposed than this city-general call. A bounded negative worth recording: Almaty's binding planning-constraint maps name seismicity, faults and soils, and contain zero occurrences of any mudflow term across 130 kB of text. | seed | seed | e75dee88-c831-4b5a-bbe3-a3d826c10aad | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 119 | 5ead5845-61b0-4fb6-a317-0ab21fef7b26 | th | english-proficiency | 402 | EF EPI score (300-700) | https://www.ef.com/assetscdn/WIBIwq6RdJvcD9bc8RMd/cefcom-epi-site/fact-sheets/2025/ef-epi-fact-sheet-thailand-english.pdf | EF Education First — EF English Proficiency Index 2025 | 2025-01-01 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | EF EPI 2025 edition, Thailand: score 402, global rank #116, against a global average of 488 — a 13-point fall on the prior edition. Skill breakdown on the same sheet: Reading 416, Listening 385, Writing 363, Speaking 377. The 2025 edition is computed from 2024 test-taker data, so the reference year is 2024 and effectiveAt is set to the edition year. Extracted from EF's own country fact sheet PDF via pdftotext (the WebFetch text layer failed). Per the brief, EF is the only source for this criterion and its licensing for commercial redistribution is UNRESOLVED — flagging that, not resolving it. Caveat on interpretation: EF EPI is a self-selected online test population, not a national sample, so it is a weak proxy for the English level of staff a school would actually hire in Bangkok or Chiang Mai; the same sheet shows Bangkok at 467, well above the national 402. SOURCE QUOTE, verbatim from the page on the retrieval date: “Global ranking: #116 Thailand EF EPI score: 402 -13 Global average score: 488” | seed | seed | f73baaa7-6e55-444a-a588-488d29338321 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 120 | 0864dddf-5510-4106-9cd1-00287a64481c | th | political-stability | 54.222 | World Bank WGI percentile rank (0-100, higher better) | https://api.worldbank.org/v2/country/THA/indicator/GOV_WGI_PV.SC?format=json&source=3&date=2022:2024 | World Bank — Worldwide Governance Indicators | 2024-01-01 | 2026-07-17T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | WGI 'Political Stability and Absence of Violence/Terrorism' governance score (0-100 scale, which is the percentile rank) for Thailand, most recent year 2024 = 54.222. Trend is mildly negative: 2022 = 56.206, 2023 = 58.381, 2024 = 54.222. Note the legacy indicator codes (PV.PER.RNK / PV.EST) are now retired and return 'The indicator was not found. It may have been deleted or archived.' — the live code is GOV_WGI_PV.SC and it requires the source=3 parameter. Anyone re-verifying this must use the new code or they will wrongly conclude the source is dead. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator":{"id":"GOV_WGI_PV.SC","value":"Political Stability - Governance score (0-100)"},"country":{"id":"TH","value":"Thailand"},"countryiso3code":"THA","date":"2024","value":54.2220045}” | seed | seed | 0e47f5f8-5028-4a72-ab8b-c3c3469e4fe0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 121 | a324c6c3-b5e1-4ed9-91a3-2074421b46c4 | th | rule-of-law | 53.256 | World Bank WGI percentile rank (0-100, higher better) | https://api.worldbank.org/v2/country/THA/indicator/GOV_WGI_RL.SC?format=json&source=3&date=2022:2024 | World Bank — Worldwide Governance Indicators | 2024-01-01 | 2026-07-17T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | WGI 'Rule of Law' governance score (0-100 percentile) for Thailand, most recent year 2024 = 53.256. Flat trend: 2022 = 52.349, 2023 = 53.598, 2024 = 53.256. Live indicator code is GOV_WGI_RL.SC with source=3; the legacy RL.PER.RNK is retired and 404s. Corroborating (not substituted): the Heritage 2025 Index country page independently states 'The overall rule of law is weak in Thailand' and scores Judicial Effectiveness at 34.6. WJP Rule of Law was deliberately NOT used — it is on the prohibited non-commercial list. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator":{"id":"GOV_WGI_RL.SC","value":"Rule of Law - Governance score (0-100)"},"country":{"id":"TH","value":"Thailand"},"countryiso3code":"THA","date":"2024","value":53.2556386}” | seed | seed | 93fbdb1c-3293-4a7e-ac1f-e57921f9179e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 122 | a3a6af42-95d8-452d-b804-4d7f26d29bac | th | corruption | 33 | Transparency International CPI score (0-100, higher better) | https://www.transparency.org/en/countries/thailand | Transparency International — Corruption Perceptions Index | 2025-01-01 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | TI CPI most recent edition: Thailand scores 33/100, ranked 116 of 182, down 1 point year on year. The page's own navigation links to /cpi/2025, so this is the 2025 edition (published early 2026); effectiveAt set to the 2025 reference year. Direction check: CPI is a cleanliness score, higher = less corrupt, so 33 is poor — below the global average and consistent with the independent WGI Control of Corruption score of 37.18 (2024) and Heritage's Government Integrity score of 38.4. Three independent measures agreeing in the 33-39 band is the useful signal here. SOURCE QUOTE, verbatim from the page on the retrieval date: “Thailand has a score of 33 this year, with a change of -1 since last year, meaning it ranks 116 out of 182 countries.” | seed | seed | bad40ee7-97f0-4d8e-9d4b-91ee7e53dc2e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 123 | ca956eab-61b7-47c4-9b7b-2c0360fa0f1a | th | expropriation-risk | 45.1 | Heritage property rights score (0-100, HIGHER IS BETTER) | https://static.heritage.org/index/pdf/2025/2025_indexofeconomicfreedom_thailand.pdf | The Heritage Foundation — 2025 Index of Economic Freedom | 2025-01-01 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | Heritage 2025 Index, Thailand Property Rights = 45.1. DIRECTION CONFIRMED: this is a rights score where higher = stronger property rights = lower expropriation risk, so 45.1 is a POOR result, not a good one — it sits below the world average, which Heritage's own narrative on the same page confirms ('The country's property rights score is below the world average'). Related scores on the same page for context: Judicial Effectiveness 34.6, Government Integrity 38.4, overall economic freedom 60.6 (world rank 84, 'MODERATELY FREE'). Heritage data compiled as of November 2024, per the page footer, so the underlying reference year is 2023-24. SOURCING NOTE: heritage.org and economicfreedom.heritage.org both return 403 to fetchers at every path tried, including their API path — the source is LIVE, not dead. I reached the identical official data via Heritage's own static PDF host (static.heritage.org), which is not behind the WAF. I did NOT substitute an aggregator; theglobaleconomy.com and knoema were available and were deliberately not used. ICSID and UNCTAD were avoided as non-commercial per the brief. SOURCE QUOTE, verbatim from the page on the retrieval date: “45.1 34.6 38.4 Property Judicial Government Rights Effectiveness Integrity The overall rule of law is weak in Thailand. The country's property rights score is below the world average; its judicial effectiveness score is below the world average; and its government integrity score is above the world average.” | seed | seed | a059efe1-e70d-40d0-9fd2-7faf5ca60396 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 124 | 949cc994-c549-46d5-9ccc-71ad7491ce80 | kz-almaty-city | seismic-hazard | 3 | hazard class 0-3 | https://www.thinkhazard.org/en/report/37206 | ThinkHazard! — Global Facility for Disaster Reduction and Recovery (GFDRR) / World Bank | 2026-07-22 | 2026-07-22T00:00:00.000Z | MULTILATERAL | CORROBORATED | DOCUMENT_EXTRACTION | NULL | THE CRITERION THIS ROW LANDS IN EXISTS BECAUSE OF THIS CITY. Both the Almaty and the Tashkent adversarial verifiers concluded independently, and without being asked, that seismic hazard had no home in the CITY criterion set — and that a national INFORM composite reading 2.8 ('low risk') for a city almost completely destroyed on 3 January 1911 and again in 1887 is the coarse screen failing exactly where it always fails. D-051 added `seismic-hazard` and funded it by cutting `disaster-exposure` 12 -> 7: the weight moves from the national composite doing the flattening to the city-specific hazard. VALUE = 3 (High), the TOP of ThinkHazard's four-class legend. VERBATIM from the earthquake layer for the Almaty administrative unit (fetch_page HTTP 200, re-fetched and CONFIRMED by the adversarial verifier): 'Hazard level: High' and 'more than a 20% chance of potentially-damaging earthquake shaking in your project area in the next 50 years'. Class mapping per D-051: Very low=0, Low=1, Medium=2, High=3. IDENTIFIER TRAP, RECORDED SO NOBODY REPEATS IT: the researcher's first attempt used admin unit id 1211, which is 'Semenawi Keih Bahri', ERITREA — a different country returning a plausible 'Medium'. The correct Almaty units are 37206 and 37157, and both agree. A hazard value read off the wrong administrative unit is the exact class of error this repo's verification loop exists to catch. CORROBORATING SECOND INSTRUMENT, recorded here and deliberately NOT used as the value: peak ground acceleration of about 0.38 g at 10% probability of exceedance in 50 years (~475-year return period, rock Vs30 800 m/s), from the SFRARR Central Asia probabilistic seismic hazard assessment (Poggi et al., NHESS 2025, CC BY 4.0), verbatim 'specifically for the city of Almaty, with PGA of around 0.38 g', with a stated inter-model spread of about +/-0.1 g. That paper reports and corroborates Silacheva et al. 2018, whose ScienceDirect primary 403'd for both agents. WHY ThinkHazard IS THE SPINE AND PGA IS NOT: the only global PGA grid is GEM, and GEM's map DATA is CC BY-NC-SA 4.0 (only the poster image is CC BY-SA) while the GEM Central Asia / EMCA model is available only under a restricted licence until it opens as CC BY-NC-SA at the end of 2026. Building the criterion on PGA would therefore be a licence trap everywhere outside Central Asia. ThinkHazard is city-scoped, one fetch, and its licence is clean — verbatim: 'Classified hazard levels are licenced under CC-BY', with no NC, SA or ND clause. USGS publishes no competing hazard product for this region and points at GEM. KAZAKHSTAN'S OWN INSTRUMENT, identified and its binding force proved, though its table is paywalled ('Демо-версия документа'): SP RK 2.03-30-2017 Appendix B is a mandatory list of settlements with hazard 'в баллах и ускорениях' — the per-city legal design acceleration — made operative for this city by SP RK 2.03-31-2020, 'Застройка территории города Алматы с учетом сейсмического микрозонирования', quoted from the approved city planning regulation on adilet.zan.kz. That regulation also yields a 3-storey / 12 m height cap south of Al-Farabi avenue and a prohibition on building the mountain-front bench — Almaty's own law treats this hazard as binding on construction, which is the strongest available corroboration of a High class. effectiveAt = the retrieval date: ThinkHazard states no vintage for its underlying hazard layer, so this dates OUR read of the tool's current published classification, not the age of the model behind it. A checker who learns the layer's vintage should replace it. | seed | seed | 3d63e074-a803-453d-9108-20c7d456f1d7 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 125 | 3dd38ec6-fa64-4399-b078-5e68c3ffd6a7 | my-jhr-forest-city | visa-access | UNRESOLVED | NULL | https://www.pmo.gov.my/en/news-en/special-financial-zone-to-be-created-in-forest-city-pm-anwar/ | Prime Minister's Office of Malaysia | 2023-08-25 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | The Special Financial Zone announcement offers 'a special income tax rate of 15 per cent for skilled workers and multiple entry visas'. That is a zone instrument appearing to grant what national law withholds — exactly D-011's override case — but the announcement names no Act, gazette order, or statute, so no override is honoured and the country's gate stands. Find the enabling instrument and this becomes a real override. | seed | seed | 99f0acf1-4437-442d-a317-3ac981676fce | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 126 | 481e369c-eac8-495f-8ac4-777b0d7c8e41 | th | currency-inflation | -0.134 | CPI %/yr | https://api.worldbank.org/v2/country/THA/indicator/FP.CPI.TOTL.ZG?format=json&date=2022:2025&per_page=50 | World Bank — World Development Indicators (Inflation, consumer prices, annual %) | 2025-01-01 | 2026-07-17T00:00:00.000Z | MULTILATERAL | UNVERIFIED | OFFICIAL_API | NULL | Thailand consumer price inflation, most recent year 2025 = -0.134% — i.e. mild DEFLATION. Series: 2022 = 6.076, 2023 = 1.229, 2024 = 0.398, 2025 = -0.134. WDI series last updated 2026-07-13, so this is fresh. TWO FLAGS. (1) RANGE: the criterion specifies 0-50, and this value is NEGATIVE, so it falls below the stated floor. I am reporting the true figure rather than clamping it to 0, because clamping would silently misrepresent the data; the engine should decide how to handle a sub-floor value. If the axis treats 'lower is better' literally, deflation will score as maximally good, which is economically wrong — persistent deflation is a symptom of weak demand, not currency strength, and Thailand has run below the Bank of Thailand's 1-3% target band. (2) SOURCE: the brief names IMF WEO as the source; I used the World Bank WDI series instead, which is also multilateral and was reachable via API. I did not fetch IMF WEO, so this is marked UNVERIFIED rather than CORROBORATED — a WEO cross-check is the outstanding step. SOURCE QUOTE, verbatim from the page on the retrieval date: “{"indicator":{"id":"FP.CPI.TOTL.ZG","value":"Inflation, consumer prices (annual %)"},"country":{"id":"TH","value":"Thailand"},"countryiso3code":"THA","date":"2025","value":-0.133634356480207}” | seed | seed | 245c83f9-ec60-4aa6-b591-b1f5f37445c2 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 127 | 1271a670-7ffa-424c-bba8-c7322f001026 | th | tax-burden | UNRESOLVED | USD/member-month | https://www.rd.go.th/english/6045.html | The Revenue Department, Kingdom of Thailand | 2024-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed 1463 and an independent agent re-fetched the source and rejected it. VERIFIER: URL resolves and both the resident definition and the full progressive rate table are verbatim on the Revenue Department page, exactly as quoted. But the quote does not establish the claimed value of 1463. The figure 1463 appears nowhere on the page. It is evidently a derived computation (presumably tax owed in some unit at some income level), and neither the income assumed, the unit, nor the arithmetic is stated in the claim or derivable from the quote. A rate table is not a tax-burden figure. The parent must either supply the derivation inputs or treat 1463 as unsourced. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: MODELLED — the rate table is sourced and quoted; the income base and FX rate are MY ASSUMPTIONS and drive the answer. Treat the number as an upper bound, not a measurement. FULL WORKING. Rate base (sourced, quoted): Thai progressive PIT, 0-150,000 THB exempt, then 5/10/15/20/25/30/35% bands, top rate 35% over 4,000,000 THB. Residency trigger (sourced, quoted): 'Resident' = residing in Thailand MORE THAN 180 days in a calendar year; a resident is taxable on Thai-source income plus foreign-source income BROUGHT INTO Thailand. Assumption 1 — income: USD 80,000/yr gross remote income. I chose this because it is the only defensible anchor in the record (BOI's own LTR Work-from-Thailand threshold); it is NOT a measured member income. Assumption 2 — FX: 36.5 THB/USD, UNSOURCED, my estimate. 80,000 x 36.5 = 2,920,000 THB. Assumption 3 — full remittance of income into Thailand in the year earned (worst case for a resident living there). Assumption 4 — NO deductions or allowances applied. I deliberately omitted the 50% expense deduction (capped 100,000 THB) and the 60,000 THB personal allowance because I could not source those figures from the Revenue Department in this pass; omitting them OVERSTATES the tax, which is why this is an upper bound. Computation on 2,920,000 THB: 150k@5%=7,500; 200k@10%=20,000; 250k@15%=37,500; 250k@20%=50,000; 1,000k@25%=250,000; 920k@30%=276,000. Total 641,000 THB/yr = USD 17,562 = USD 1,463/member-month. THE KNIFE-EDGE THAT MATTERS MOST: residency requires MORE THAN 180 days, and a single DTV entry grants EXACTLY 180 days. A member on one 180-day DTV entry is therefore NOT a Thai tax resident and owes ZERO Thai tax on foreign-source income — so the true value for this criterion is bimodal, roughly 0 at exactly 180 days and ~1,463 once the discretionary +180 extension is taken and the member crosses the threshold. The scenario's '>=180 days' requirement sits precisely on this line. I report the resident case because a school cohort realistically exceeds 180 days. POLICY RISK, NOT INFERRED UPWARD: Revenue Departmental Instruction Por. 161/2566 (in force 1 Jan 2024) taxes foreign income remitted in ANY later year, not just the year earned. A Royal Decree exempting income remitted within two tax years has been DRAFTED but NOT ENACTED and is stalled pending elections — per the brief's rule that a government intention is not a statute, I modelled the enacted regime, not the proposal. QUOTE THE RESEARCHER OFFERED: “"Resident" means any person residing in Thailand for a period or periods aggregating more than 180 days in any tax (calendar) year. A resident of Thailand is liable to pay tax on income from sources in Thailand as well as on the portion of income from foreign sources that is brought into Thailand. Taxable Income (baht) | Tax Rate (%) 0-150,000 | Exempt more than 150,000 but less than 300,000 | 5 more than 300,000 but less than 500,000 | 10 more than 500,000 but less than 750,000 | 15 more than 750,000 but less than 1,000,000 | 20 more than 1,000,000 but less than 2,000,000 | 25 more than 2,000,000 but less than 4,000,000 | 30 Over 4,000,000 | 35” | seed | seed | bac74e0f-fc36-4315-96c0-01630c38eab7 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 128 | de15eb8e-2486-4ef6-827d-0050bd5a2592 | th | import-duty | UNRESOLVED | USD | https://www.wto.org/english/res_e/statis_e/daily_update_e/tariff_profiles/TH_e.pdf | World Trade Organization — Tariff Profile: Thailand | 2025-01-01 | 2026-07-17T00:00:00.000Z | MULTILATERAL | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | Not resolved, because the two terms that determine the answer are both missing and only one is close. What I DID source (quoted): the WTO Tariff Profile for Thailand gives, for the product group 'Mechanical, office and computing machinery', an MFN applied average of 2.1% with 68.9% of tariff lines duty-free (2025). That is a GROUP AVERAGE across a broad category, NOT the specific tariff line the criterion asks me to name. What is MISSING: (1) the specific line — I could not retrieve Thailand's applied rate for HS 8471 (automatic data processing machines) from Thai Customs; the Integrated Tariff Database at itd.customs.go.th timed out, customs.go.th returns 200 but serves the tariff lookup through JavaScript that a fetcher cannot execute, and tariff-finder.asean.org timed out. (2) the equipment basket — the criterion wants a one-off USD duty, which requires a per-member equipment spend that appears nowhere in my brief and that I refuse to invent, since it is the dominant term in the answer. Every source asserting 'laptops are duty-free in Thailand' that I found was a freight forwarder, a customs broker or a duty-calculator site — commercial listings, not the tariff schedule. The claim is very likely true (Thailand is an ITA signatory and the WTO group data shows ~69% of lines in this category duty-free), but 'very likely true' with a real URL attached is exactly the failure mode the brief warns about, so it stays UNRESOLVED. To close: get the HS 8471.30 applied rate from the Thai Customs Integrated Tariff Database via a browser, and get a per-member equipment basket from the property/city pass. SOURCE QUOTE, verbatim from the page on the retrieval date: “Mechanical, office and computing machinery 19.1 10.3 30 92.4 2.1 68.9 30 11.6 66.7” | seed | seed | faef1632-7922-498f-b9c1-be9fc8456157 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 129 | 49bc10cc-9910-4ae4-9972-e522f3407cd7 | th | entry-friction | UNRESOLVED | days | https://www.thaievisa.go.th/visa/dtv-visa | Official Website of Thailand Electronic Visa (Ministry of Foreign Affairs) | 2025-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | No official processing-time commitment could be sourced, so no cohort-weighted figure can be computed. The official e-Visa portal (thaievisa.go.th) — which has been the mandatory channel for all DTV applications since 1 Jan 2025 — is a JavaScript single-page application; both the /visa/dtv-visa page and /visa-types render nothing but the header 'Official Website of Thailand Electronic Visa' to a fetcher, hence the empty quote. The MFA checklist PDF and the Royal Thai Embassy post pages state required documents and fees but commit to no processing time. EVERY processing-time figure I found (variously '5 to 15 business days', '3 to 7 days', '10 to 20 business days') came from relocation agencies, visa-service firms or Quora — precisely the category the brief bars for visa content, since these firms profit from optimism about speed. I will not average marketing claims into a number. One genuine official datapoint exists but does not answer this criterion: Private School Act s.121 gives the Licensor ninety days to decide a school licence application — that is the school's licence, not a member's visa. To close: drive the e-Visa portal with a headless browser, or obtain a published service standard from a Royal Thai Embassy post. | seed | seed | ab8f3afb-db4e-41a1-947b-909e08228029 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 130 | 5c47f31f-9d63-4b97-8473-196afaa1a18e | th | government-receptivity | 3 | rubric 0-5 | https://www.boi.go.th/upload/content/BOI_A_Guide_EN.pdf | Thailand Board of Investment (BOI) | 2026-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 3: Named policy fit: a programme this project class demonstrably qualifies for, published eligibility criteria (SEZ regime, education-hub licence, digital-nomad statute) THE FACTS: · BOI's current (2026) investment promotion guide lists a named promoted activity for higher-education campuses, activity 10.7.2, at the A1 incentive tier, with published eligibility conditions — a project class fit for a Network School node. https://www.boi.go.th/upload/content/BOI_A_Guide_EN.pdf “10.7.2 High-potential academies and institutions for higher education” · The published eligibility criteria for that activity require an EEC or designated SEZ location, and require foreign-invested projects to be approved by a named commission. https://www.boi.go.th/upload/content/BOI_A_Guide_EN.pdf “Project must be located in the Eastern Economic Corridor (EEC) or other special economic zones for high-potential academies designated by the Cabinet.” · Foreign investment in the higher-education activity is gated on approval by the Commission for High-potential Foreign Higher Education Institutions — published criteria, not open engagement. https://www.boi.go.th/upload/content/BOI_A_Guide_EN.pdf “Project must be high-potential institutions for higher education. In case of foreign investment, projects must be approved by the Commission for High-potential Foreign Higher Education Institutions.” · Thailand additionally operates a digital-nomad statute (the DTV) with published eligibility criteria that the node's member cohort qualifies for — the rubric's own example of a named policy fit. https://image.mfa.go.th/mfa/0/RzaiZWKBzF/consular/Visa/18.Destination_Thailand_Visa_(DTV).pdf “Workcation (digital nomad/ remote worker/ foreign talent/ freelancer)” FACTS: BOI's 2026 promotion guide carries activity 10.7.2, 'High-potential academies and institutions for higher education', at A1 incentives, with published eligibility conditions (EEC/SEZ location; foreign-invested projects approved by the Commission for High-potential Foreign Higher Education Institutions), and 10.7.1 vocational training centres alongside it. Thailand separately runs the DTV, a published digital-nomad route the member cohort qualifies for. MAPPING: anchor 3 — 'a programme this project class demonstrably qualifies for, published eligibility criteria', whose parenthetical names both an SEZ regime and a digital-nomad statute. WHAT ARGUED HIGHER: anchor 4 needs a named empowered official or agency to have met, corresponded, or issued a public statement about THIS project or a close analogue, on the record. I searched for any Thai official engagement with a network state / network school / charter city and found nothing; the BOI regime names an activity class, not this project, and an A1 incentive listing in a guidebook is a published schedule, not engagement. Anchor 5 needs a signed MOU, gazetted incentive or allocated land naming the project or its class — none exists. WHAT WOULD MOVE IT: down to 2 if the BOI education activity were shown withdrawn or inapplicable, leaving only generic FDI welcome; up to 4 on a sourced, on-the-record statement by a named BOI/EEC official about this project or a close analogue; up to 5 on a signed instrument. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: Re-fetched both PDFs. The BOI guide is image-heavy and defeats WebFetch, so I extracted it with pdftotext: p.192 of the 'Investment Promotion Guide 2026' carries '10.7.2 High-potential academies and institutions for higher education' at the A1 tier, alongside '10.7.1 Vocational training center'. Both cited conditions appear verbatim: 'Project must be located in the Eastern Economic Corridor (EEC) or other special economic zones for high-potential academies designated by the Cabinet' and 'In case of foreign investment, projects must be approved by the Commission for High-potential Foreign Higher Education Institutions'. The DTV PDF carries 'Workcation (digital nomad/ remote worker/ foreign talent/ freelancer)'. MAPPING HOLDS at anchor 3. The DTV alone is dispositive: anchor 3's own parenthetical names a 'digital-nomad statute' with published eligibility criteria, and that is exactly what the DTV is. The BOI leg is weaker than the note implies — 10.7.2 requires a focus on science and technology education, an EEC/SEZ location, and Commission approval, so a Network School node's qualification is arguable rather than 'demonstrable' — but anchor 3 does not depend on it. I checked the upward path and agree it is closed: no named official engagement about this project or a close analogue (anchor 4), no instrument naming the project or its class (anchor 5). Downward is also closed: BOI 10.7.2 names an activity class and the DTV names a cohort, so this is not anchor 2's 'generic openness only'. Correctly refused to infer upward from an A1 listing to engagement. | seed | seed | e4fd13c2-e84c-442a-b8c9-ab54f1ad6885 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 131 | a6680219-a807-437e-945e-05d7d442b55c | th | visa-preference | 3 | rubric 0-5 | https://image.mfa.go.th/mfa/0/RzaiZWKBzF/consular/Visa/18.Destination_Thailand_Visa_(DTV).pdf | Ministry of Foreign Affairs, Kingdom of Thailand | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 3: A preferential route exists and fits, but is discretionary, capped, pilot-stage, or not in force THE FACTS: · The Destination Thailand Visa is a preferential route expressly for remote work, published by the MFA with multi-year terms. https://image.mfa.go.th/mfa/0/RzaiZWKBzF/consular/Visa/18.Destination_Thailand_Visa_(DTV).pdf “Workcation (digital nomad/ remote worker/ foreign talent/ freelancer)” · The DTV is multiple-entry, valid five years from issue, with 180 days of stay per entry. https://image.mfa.go.th/mfa/0/RzaiZWKBzF/consular/Visa/18.Destination_Thailand_Visa_(DTV).pdf “180 days each time” · Published financial eligibility criterion for the DTV. https://image.mfa.go.th/mfa/0/n3gTFT2TOE/Visa_Requirements/Checklist_DTV.pdf “A copy of the bank statement for the last 3 months with an ending balance of no less than 500,000 THB or (65,000 MYR).” · The issuing posts publish an express reservation of discretion to refuse the DTV — it is not available as of right on the published criteria. https://hochiminh.thaiembassy.org/en/page/destination-thailand-visa-dtv-th%E1%BB%8B-th%E1%BB%B1c-%C4%91i%E1%BB%83m-%C4%91%E1%BA%BFn-th “The Royal Thai Consulate-General in Ho Chi Minh reserves the rights to request additional documents, interviews, or to disapprove as deemed necessary.” · A second issuing post publishes the same reservation of consular discretion, showing it is the regime's standing posture and not one post's wording. https://yangon.thaiembassy.org/en/publicservice/destination-thailand-visa-dtv “The consular officer reserves the right to request additional documents and do an interview as deemed necessary.” · The MFA checklist imposes extra mandatory documentation on a named list of specific countries, so the route is not uniform across cohorts. https://image.mfa.go.th/mfa/0/n3gTFT2TOE/Visa_Requirements/Checklist_DTV.pdf “Additional Documents for SPECIFIC COUNTIRES (please click to check list of countries):” FACTS: the DTV exists, is in force, is published by the MFA with multi-year terms (multiple entry, 5 years from issue, 180 days per entry, extendable once), and is expressly built for the activity class — 'Workcation (digital nomad/ remote worker/ foreign talent/ freelancer)' — on published criteria including a 500,000 THB balance. Against that, the issuing posts publish an express reservation of discretion to refuse: Ho Chi Minh 'reserves the rights to request additional documents, interviews, or to disapprove as deemed necessary', and Yangon publishes the same. The MFA checklist also layers mandatory extra documents on a named list of specific countries. MAPPING: anchor 3 — 'A preferential route exists and fits, but is discretionary'. WHAT ARGUED HIGHER: this is the strongest case on the board for 5 — the DTV is a remote-work route the project squarely fits, with published multi-year terms, which is anchor 5's own description. But anchor 5 is defined as 'As 4, AND designed for this activity class', and anchor 4 requires the route be 'open AS OF RIGHT on published criteria'. The issuing authority publishes, in terms, that it may disapprove as deemed necessary. That is a reservation of discretion by the body that grants the thing, not consular boilerplate about processing times, so anchor 4's gate does not open and 5 cannot be reached through it. I also could not confirm from a primary that all four cohorts (US/EU/IN/CN) are covered on equal terms: US and EU posts publish DTV pages, the Nanning consulate publishes a DTV page in Chinese (image-only, unquotable) and New Delhi confirms e-Visa operation in India from 1 Jan 2025, but the 'SPECIFIC COUNTIRES' list means terms are not uniform. WHAT WOULD MOVE IT: up to 4 or 5 on a primary showing the DTV is granted on criteria without a reserved discretion to refuse, plus a per-nationality eligibility list covering US/EU/IN/CN; down to 2 if a cohort were shown excluded, or if the 180-day entry proved not to cover the intended activity. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: All six quotes verify verbatim. The DTV PDF yields 'Workcation (digital nomad/ remote worker/ foreign talent/ freelancer)', '180 days each time', and 'Multiple | 5 years from issue date'. The Checklist PDF (extracted via pdftotext) yields 'A copy of the bank statement for the last 3 months with an ending balance of no less than 500,000 THB or (65,000 MYR)' and 'Additional Documents for SPECIFIC COUNTIRES (please click to check list of countries):' — typo included. Ho Chi Minh publishes 'The Royal Thai Consulate-General in Ho Chi Minh reserves the rights to request additional documents, interviews, or to disapprove as deemed necessary.' Yangon publishes 'The consular officer reserves the right to request additional documents and do an interview as deemed necessary.' TWO PROVENANCE CAVEATS the note does not surface, neither fatal: (1) the MFA-hosted DTV PDF is the JAPAN post's version (52,000 yen fee, My Number Card, TOKIBOTOHON, 戸籍謄本) and the Checklist is the MALAYSIA post's (MYR, Malaysian PR card, Wisma Putra) — these are per-post documents on the MFA domain, not a single national instrument, which if anything strengthens the non-uniformity finding. (2) Yangon's reservation is NOT 'the same' as Ho Chi Minh's — it omits the 'or to disapprove' limb, so only one post is sourced for a reserved power to refuse. MAPPING HOLDS at anchor 3. Anchor 5 is reachable only through anchor 4, and anchor 4 demands the route be 'open AS OF RIGHT on published criteria'; the issuing authority's published reservation to 'disapprove as deemed necessary' defeats that, which is anchor 3's 'discretionary' in terms. This is the conservative read and it is the correct one — the note resisted a genuinely tempting 5. | seed | seed | 5c363513-5c98-42d5-b2c7-173b22812bb4 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 132 | 6528b480-84bb-4f19-a498-d895043455c7 | th | web3-support | 4 | rubric 0-5 | https://www.nationthailand.com/news/policy/40055134 | The Nation Thailand | 2025-09-05 | 2026-07-17T00:00:00.000Z | REPUTABLE_MEDIA | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: Framework PLUS published tax treatment PLUS documented public-sector engagement THE FACTS: · A licensing framework exists: Thai tax law defines its scope by reference to licensed digital asset exchanges, brokers and dealers under the digital asset business law. https://www.nationthailand.com/news/policy/40055134 “399th ministerial regulation” · Published tax treatment: a ministerial regulation signed by a named deputy finance minister and published in the Royal Gazette exempts capital gains tax on digital asset trading through licensed operators for a defined five-year window. https://www.nationthailand.com/news/policy/40055134 “From 1 January 2025 to 31 December 2029” · Banking: the Bank of Thailand has a named notification regulating commercial banks' financial business groups undertaking digital asset businesses — i.e. Thai banking groups may lawfully operate in the sector under BOT supervision. https://www.bot.or.th/en/financial-innovation/resilient-regulation/digital-asset.html “Notification of the Bank of Thailand No. FPG. 6/2565 Re: Regulations on commercial banks' financial business groups undertaking digital asset related businesses and transactions” · Documented public-sector engagement: the SEC and BOT jointly developed a sandbox letting tourists convert digital assets to baht for domestic spending, with published spending limits. https://www.nationthailand.com/business/digital-assets/40052606 “exchange digital assets for Thai baht and use them for spending within the country” · Countervailing: digital assets are not accepted as a means of payment — the sandbox requires conversion to baht first, and the BOT/SEC/MoF issued joint guidelines against digital assets as a means of payment. https://www.bot.or.th/en/financial-innovation/resilient-regulation/digital-asset.html “Notification of the Bank of Thailand No. FPG. 6/2565 Re: Regulations on commercial banks' financial business groups undertaking digital asset related businesses and transactions” FACTS: (a) Framework — a licensing regime under the Emergency Decree on Digital Asset Businesses, evidenced here by Thai tax law defining its own scope by reference to licensed exchanges/brokers/dealers; (b) published tax treatment — ministerial regulation No. 399, signed by Deputy Finance Minister Julapun Amornvivat and published in the Royal Gazette, exempting capital gains tax on digital asset trading via licensed operators 'From 1 January 2025 to 31 December 2029'; (c) documented public-sector engagement — the SEC and BOT jointly ran a public consultation and sandbox (TouristDigiPay) for converting digital assets to baht for spending, with published per-merchant limits. BANKING (the rubric's note, DATA-SOURCES §194): not skipped — BOT Notification No. FPG. 6/2565 regulates commercial banks' financial business groups undertaking digital asset related businesses, so bank-group participation is permitted and supervised on the record. I did NOT find a primary establishing that digital-asset firms routinely obtain deposit accounts in practice; that specific fact remains unsourced and is the weakest leg of this score. MAPPING: anchor 4. WHAT ARGUED HIGHER: anchor 5 lists 'a named government programme, sandbox' and TouristDigiPay is exactly that, which is a real case for 5. It did not reach because (i) my TouristDigiPay evidence is media-only — the SEC's own page for it returned HTTP 403 to every attempt (see blocks below), so I cannot quote the issuer or confirm from a primary that the trial actually launched and is running as of 2026-07-17; and (ii) anchor 5 reads 'Active public-sector adoption ... or acceptance of digital assets', and Thailand does not accept digital assets — the sandbox exists precisely because tourists must convert to baht before spending, and the BOT/SEC/MoF have joint guidelines against digital assets as a means of payment. Adoption of a conversion rail is not acceptance of the asset. BLOCKS TO REPORT: the entire sec.or.th domain returns 403 to this agent — https://www.sec.or.th/EN/pages/lawandregulations/digitalassetbusiness.aspx, https://www.sec.or.th/EN/Pages/Shortcut/TouristDigiPay.aspx and two other paths all 403 via WebFetch and via curl with a browser UA. The SEC licensed-operator register is therefore unquoted. Bangkok Post returned HTTP 402 (paywall). WHAT WOULD MOVE IT: up to 5 on an SEC/BOT primary confirming TouristDigiPay is live and operating; down to 3 if the tax exemption lapsed or if banking access for digital-asset businesses were shown unobtainable in practice. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: Load-bearing facts verify. Nation Thailand confirms the '399th ministerial regulation', signed by then deputy finance minister Julapun Amornvivat on 27 August, published in the Royal Gazette 5 September, exempting capital gains on transfers 'through a digital asset exchange, through a licensed digital asset broker, or transferred to a licensed digital asset dealer' for income received 'from 1 January 2025 to 31 December 2029'. The BOT page carries 'Notification of the Bank of Thailand No. FPG. 6/2565' (issued 6 Oct 2022). TWO DEFECTS I am recording but which do not degrade the score, because both cut AGAINST it rather than propping it up: (1) fact 5's quote is a copy-paste duplicate of fact 3's FPG 6/2565 title and does NOT establish the means-of-payment claim — though the same page independently does, carrying 'BOT, SEC, and MOF to jointly set guidelines on the usage of digital assets as a means of payment for goods and services'; (2) fact 4's claim that SEC and BOT 'jointly developed a sandbox' overstates the source — the article is dated 16 July 2025 and describes a PROPOSAL under public consultation open until 13 August 2025, and never uses the name 'TouristDigiPay'. The note concedes the launch is unconfirmed. MAPPING HOLDS at anchor 4. Framework: yes. Published tax treatment: yes, Royal Gazette, unimpeachable. Documented public-sector engagement: yes — a joint SEC/BOT published proposal with per-merchant limits (50,000 and 500,000 baht/person/month) is engagement, a lower bar than adoption, and survives the corrected 'proposed' reading. On the rubric's BANKING note: not skipped, but FPG 6/2565 regulates BANKS' groups doing digital-asset business, which is not the same proposition as digital-asset firms obtaining accounts; the note says so itself. I did not degrade on this because doubting banking would also break anchor 3 (which requires it) and push toward anchor 2's 'no licensing regime', which is plainly false. Anchor 5 correctly refused: a conversion rail is not acceptance of the asset. | seed | seed | 21da15d2-6de5-4445-9138-70f898039e6d | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 133 | 0ff2140b-a1b3-4014-a878-55ad3bf6d9b1 | th | capital-controls | 3 | rubric 0-5 | https://www.bot.or.th/en/our-roles/financial-markets/foreign-exchange-regulations/exchange-control-regulation.html | Bank of Thailand | 2026-07-17 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | UNVERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 1: Reporting or registration only; no approval needed; no documented delays THE FACTS: · Repatriation of income — dividends, profits, service fees, royalties — is permitted without approval, up to the amount of the underlying obligation. https://www.bot.or.th/en/our-roles/financial-markets/foreign-exchange-regulations/exchange-control-regulation.html “Payments for goods and services such as service fees, interest, dividends, profits, or royalties, or payments related to education, are allowed up to the amount of the obligations.” · Large outward transfers require supporting documents, not approval — a reporting requirement with a published threshold. https://www.bot.or.th/en/our-roles/financial-markets/foreign-exchange-regulations/exchange-control-regulation.html “For transactions in an amount equivalent to USD 200,000 or above, the authorized banks need to request the customers to submit supporting documents except for the case where the authorized banks have performed the Know Your Business (KYB) process on the customers.” · A repatriation/surrender obligation applies to large foreign currency receipts, satisfiable by deposit into a foreign currency account at an authorized bank. https://www.bot.or.th/en/our-roles/financial-markets/foreign-exchange-regulations/exchange-control-regulation.html “Any person receiving foreign currencies from abroad in an amount equivalent to USD 1 million or above is required to repatriate such funds immediately within 360 days from the export date or the transaction date, and sell to an authorized bank or deposit them in a foreign currency account with an authorized bank within 360 days of receipt.” · A named category of transfers does require prior BOT approval: payments associated with digital assets. https://www.bot.or.th/en/our-roles/financial-markets/foreign-exchange-regulations/exchange-control-regulation.html “payments associated with digital assets” REMAPPED BY ADVERSARIAL VERIFICATION (D-032/D-035). The scorer claimed 1 at "anchor 1: Reporting or registration only; no approval needed; no documented delays". An independent agent re-checked the facts against the pages AND the mapping against contracts/rubric.md, and the facts do not reach that anchor. VERIFIER: All four quotes verify on the BOT page, including the one that sinks the score: 'payments associated with digital assets' is listed as REQUIRING PRIOR BANK OF THAILAND APPROVAL. Source discipline is properly disclosed (AREAER not used, BOT substituted, said so as the rubric demands), so that is not the defect. MAPPING FAILS. Anchor 1 reads 'Reporting or registration only; NO APPROVAL NEEDED; no documented delays.' The note's own fact 4 documents a standing prior-approval requirement for a named category, and the note itself concedes it is 'directly relevant to a node with a digital-asset treasury' and 'pointing at anchor 2 or 3' — then scores 1, BELOW both anchors it just identified. That is a self-contradiction on the face of the note. The stated escape does not work: reasoning that no published source shows whether approval is 'routinely granted, on what criteria, or how long it takes' defeats ANCHOR 2, whose mitigating limb is 'routinely granted on published criteria' — it does not license a fall back to anchor 1, whose condition is that no approval is needed at all. Absence of evidence that approval is easy is not evidence that approval is not required; the requirement is documented on the regulator's own page. CORRECTED TO ANCHOR 3, 'Approval discretionary': approval is required for a named category central to this node's normal operation, and with no published criteria constraining the BOT's grant, it is discretionary by default. Anchor 2 is unavailable precisely because its 'routinely granted on published criteria' limb is unsourced. This is the lower-is-better analogue of inferring upward — the score reads as more favourable than the scorer's own evidence supports, on a subject that hard-caps the Risk axis. ORIGINAL SCORING NOTE, retained — its facts stand and only the mapping moved: SOURCE DISCIPLINE, stated as the rubric requires: the rubric names IMF AREAER specifically and I did NOT use it. elibrary-areaer.imf.org redirects to an IMF bookstore auth gateway (302) and is not readable without a subscription, so I reasoned from the Bank of Thailand's own exchange-control regulation instead. That substitution is the main weakness of this score and is why I marked it UNVERIFIED. FACTS: income and capital repatriation is permitted without approval up to the amount of the obligation; outward transfers at or above USD 200,000 require supporting documents to the authorized bank (or prior KYB), which is reporting, not approval; receipts from abroad of USD 1m or above must be repatriated within 360 days, satisfiable by deposit into an FCD at an authorized bank; a USD 50,000 per person per year cap applies to gifts/grants but not to transfers to emigrants or family. MAPPING: anchor 1 — 'Reporting or registration only; no approval needed'. WHAT ARGUED HIGHER (worse): three things. (1) The BOT does require prior approval for a named list including 'payments associated with digital assets' — directly relevant to a node with a digital-asset treasury, and pointing at anchor 2 or 3. It did not carry the score because anchor 2 is framed as approval required for LARGE transfers and anchor 3 turns on approval being discretionary or delayed; the digital-asset approval is a category rule, and I found no published source on whether it is routinely granted, on what criteria, or how long it takes. (2) The USD 1m repatriation obligation is a genuine control, but it constrains where funds sit, not whether capital can leave, and is satisfiable by an onshore FCD. (3) Anchor 0 is excluded outright — these are controls, so 'No controls' is false. WHAT WOULD MOVE IT: up to 2 or 3 on an AREAER entry or BOT source showing approval required for ordinary large repatriations, or documented delays / discretionary refusal — particularly on the digital-asset approval; down to 0 only if the documentation threshold and the USD 1m rule were removed. Given the hard-cap role of this subject on the Risk axis, an AREAER re-check should be treated as required before this score is relied on. | seed | seed | 066736b0-3196-4e25-8654-59926bb5ab01 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 134 | 3a48bc01-43b0-4191-9de0-baf0ebdbfab4 | th | policy-reversal-precedent | 4 | rubric 0-5 | https://mfa.go.th/en/content/summary-press-briefing-190526?cate=674059168d902f116b4a9a52 | Ministry of Foreign Affairs, Kingdom of Thailand | 2026-05-19 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A programme closed or suspended to new entrants with little notice THE FACTS: · THE INSTANCE: on 19 May 2026 the Thai Cabinet revoked, in full, the 60-day visa exemption scheme it had granted to 93 countries and territories in July 2024. https://mfa.go.th/en/content/summary-press-briefing-190526?cate=674059168d902f116b4a9a52 “revoking the 60-day visa exemption scheme for all 93 countries/territories” · The same decision cut the Visa on Arrival scheme from 31 eligible countries to 4, removing access entirely for 27 nationalities. https://mfa.go.th/en/content/summary-press-briefing-190526?cate=674059168d902f116b4a9a52 “revising Visa on Arrival (VoA) and reducing the list of eligible countries/territories from 31 to 4 countries/territories” · The notice period is 15 days from Royal Gazette publication of the Ministry of Interior announcements. https://mfa.go.th/en/content/summary-press-briefing-190526?cate=674059168d902f116b4a9a52 “15 days after publication” · The replacement 30-day scheme is narrower than what it replaces, both in scope (tourism only) and in the number of eligible countries. https://mfa.go.th/en/content/summary-press-briefing-190526?cate=674059168d902f116b4a9a52 “introducing the new 15-day visa exemption scheme for tourism purpose for 3 countries/territories” CITED INSTANCE, as the rubric demands: on 19 May 2026 the Thai Cabinet approved 'revoking the 60-day visa exemption scheme for all 93 countries/territories' — a visa regime this same state had granted in July 2024 — and 'revising Visa on Arrival (VoA) and reducing the list of eligible countries/territories from 31 to 4', stripping VoA from 27 nationalities. The measures take effect '15 days after publication' in the Royal Gazette. This is the state revoking a visa regime it previously granted, inside two years of granting it. MAPPING: anchor 4 — 'A programme closed or suspended to new entrants with little notice'. The 60-day scheme is closed to everyone on 15 days' notice; that is little notice on any reading. WHAT ARGUED HIGHER: anchor 5 is 'Grants revoked from existing holders, or a regime abolished without transition', and the 60-day regime IS abolished outright, which is a real argument for 5. It did not reach for two reasons. First, a visa exemption is not a grant held by an identifiable holder — it is an entitlement conferred at entry — so 'grants revoked from existing holders' does not describe it; anchor 5 would require, say, revoking DTVs or BOI incentives already issued, which this decision does not do. Second, there is a transition of sorts: 15 days' notice, and most affected nationals fall back to a replacement 30-day scheme rather than to nothing, so 'abolished without transition' overstates it. WHAT ARGUED LOWER: anchor 3 ('tightened AND applied to existing holders, with a transition period') understates it — this is revocation of a scheme and total removal of VoA for 27 countries, not a tightening. I also could not verify from any primary whether travellers already in Thailand keep their existing permitted stay; the MFA briefing is silent on transitional arrangements, and the Bangkok Post report that would have covered it returned HTTP 402 (paywall) and its tollbit redirect also 402. Note this instance concerns the ordinary visa-exemption route, not the DTV or BOI regimes the project would actually use — no reversal of those is on record — but the rubric asks whether this STATE has revoked an incentive, licence or visa regime it previously granted, and it has. WHAT WOULD MOVE IT: down to 2 on a primary showing existing entitlements grandfathered and the change purely prospective; up to 5 on evidence that issued visas or granted BOI incentives were themselves revoked. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: The 19 May 2026 MFA briefing verifies: '(2) revoking the 60-day visa exemption scheme for all 93 countries/territories', '(5) revising Visa on Arrival (VoA) and reducing the list of eligible countries/territories (from 31) to 4', 'will take effect 15 days after publication', and '(4) introducing the new 15-day visa exemption scheme for tourism purpose for 3 countries/territories'. Minor quote/claim slippage on fact 4: the cited quote concerns the 15-day/3-country scheme, not the 30-day scheme the claim describes — but the page independently confirms a revised 30-day tourism exemption for 54 countries, so the claim (narrower in scope and country count than the 93-country scheme it replaces) stands. MAPPING HOLDS at anchor 4. Note this axis is LOWER IS BETTER, so the generosity risk runs toward under-scoring, and the note does not take it: it explicitly rejects anchor 3 as understating a revocation. Anchor 5's two limbs both fail on their own words — 'grants revoked from existing holders' does not describe a visa exemption, which confers an entitlement at entry rather than a held grant; and 'a regime abolished without transition' is defeated by the 15 days' notice plus a fallback 30-day scheme catching 54 of the 93 affected countries. Anchor 4 slightly understates the facts ('closed to new entrants' vs. outright revocation), but it is the closest anchor whose gate actually opens, and the note is candid that the reversal touches the ordinary exemption route rather than the DTV or BOI regimes the project would use. | seed | seed | a9945e65-176f-46db-82e7-67fbf4e9fbb5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 135 | 26950f9f-1de9-4bcc-8bc7-b5bd165e1056 | uz | visa-access | {"route":"Visa-free entry for U.S. citizens (30-day, tourism/short business)","permits":["VISIT"],"maxStayDays":30,"passportSet":"US"} | NULL | https://kun.uz/en/news/2025/12/24/uzbekistan-introduces-30-day-visa-free-entry-for-us-citizens-from-january-1 | Kun.uz (reporting a Presidential Decree of the Republic of Uzbekistan) | 2026-01-01 | 2026-07-19T00:00:00.000Z | REPUTABLE_MEDIA | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SHORT-STAY VISIT ONLY — DOES NOT QUALIFY. Fails the scenario on duration (30 < 180) and on activity (no RESIDE beyond 30 days, no STUDY, no WORK_REMOTE). SOURCE QUOTE, verbatim: "Under a presidential decree, citizens of the United States will be allowed to enter Uzbekistan without a visa for stays of up to 30 days starting January 1, 2026... The decree, signed on November 3, 2025, introduces a 30-day visa-free regime for U.S. nationals entering Uzbekistan... U.S. citizens will be permitted to remain in the country without a visa for up to 30 days from the date of entry." PURPOSE is tourism/short business, verbatim: "The measure is aimed at creating more favorable conditions for the development of Uzbek-American relations, including trade and economic cooperation, cultural and humanitarian ties, and tourism." No work or study grant is stated; the decree "does not exempt travelers from complying with migration regulations" (temporary-residence registration still required). TIER DISCLOSURE: this is REPUTABLE_MEDIA, not the decree itself — I could not retrieve the primary Presidential Decree text (lex.uz / president.uz not fetched in this pass); the decree number is not given in the article. ALTERNATIVE PRIMARY ROUTE for the same cohort: the MFA visa page (https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi, OFFICIAL_PRIMARY, last updated 2025-03-06, i.e. BEFORE this decree) lists the United States in Appendix 1 as an e-visa country (single/double/multiple e-visa "for up to 30 days") and, in Note 5, grants visa-free 30-day tourism only to U.S. citizens "who have reached the age of 55" — so before 1 Jan 2026 the general-age U.S. route was the 30-day e-visa; from 1 Jan 2026 it is 30-day visa-free. Either way the U.S. short-stay ceiling is 30 days and VISIT-only. PER-ENTRY-STAY: 30 days is the per-entry stay from date of entry, quoted above. | seed | seed | dbfac2c5-8b8c-4922-8d9f-3990544c9ff1 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 136 | ba1f3210-6df0-48ee-a87c-97983e181620 | uz | visa-access | {"route":"Visa-free entry (30-day) — bilateral/decree visa-free list","permits":["VISIT"],"maxStayDays":30,"passportSet":"DE"} | NULL | https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi | Ministry of Foreign Affairs of the Republic of Uzbekistan | 2019-02-01 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SHORT-STAY VISIT ONLY — DOES NOT QUALIFY. Germany appears as item 58 ("Federal Republic of Germany") in the MFA's "Up to 30 days" visa-free list. Keyed per-state "DE" and NOT under a coined "EU/EEA/CH" bloc, per PASSPORT.md: the MFA instrument ENUMERATES each member state individually (it does not state an EU/EEA/CH bloc), so coining a bloc key here would invent a source the page does not contain and could over-admit. NOTE ON GATE MATCHING: because the scenario's EU cohort is keyed "EU/EEA/CH" and this claim is keyed "DE", they will not string-match and the EU cohort will resolve UNKNOWN for this route — which is the safe-but-silent behaviour PASSPORT.md §1 describes, and is moot here because the route is non-qualifying anyway. THE FULL ENUMERATED EU/EEA/CH SET on the MFA 30-day visa-free list, verbatim from the page: Austria (12), Vatican (15), Luxembourg (16), Hungary (17), Greece (20), Ireland (21), Iceland (22), Italy (23), Andorra (25), Liechtenstein (26), Monaco (27), Belgium (28), Denmark (29), Spain (30), Netherlands (31), Norway (32), Sweden (33), Latvia (34), Lithuania (35), Malta (45), Poland (46), San Marino (47), Slovenia (49), Croatia (50), Portugal (40), Bulgaria (41), Cyprus (43), Romania (52), Germany (58), Finland (59), France (60), Czech Republic (62), Swiss Confederation (63), Estonia (64) [Slovak Republic (54) and United Kingdom (55) also listed]. Duration rule, verbatim: "The 30-day and 60-day visa-free regime is valid from the moment of crossing the state border of the Republic of Uzbekistan, and upon its expiration, the foreign citizen is obliged to leave the republic" and "If it is necessary to stay in the Republic of Uzbekistan for more than 30 or 60 days, a foreign citizen must provide the necessary documents for issuing one of the categories of entry visas." So maxStayDays 30 is the per-entry stay, and STUDY/WORK_REMOTE/long RESIDE are NOT granted — a longer stay requires switching to a visa category. effectiveAt 2019-02-01 is the widely reported date EU-state visa exemption began; the MFA page itself states no date, so the date is from secondary reporting and is UNVERIFIED against the primary decree. | seed | seed | cb323b11-51a8-4c0b-8739-3e848babb238 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 137 | 05be6de6-2a42-4a84-b3df-4ecdbbd8a2e9 | uz | visa-access | {"route":"Visa-free entry (30-day) — bilateral/decree visa-free list","permits":["VISIT"],"maxStayDays":30,"passportSet":"FR"} | NULL | https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi | Ministry of Foreign Affairs of the Republic of Uzbekistan | 2019-02-01 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SHORT-STAY VISIT ONLY — DOES NOT QUALIFY. France appears as item 60 ("French Republic") in the MFA's "Up to 30 days" visa-free list. Recorded as a SECOND representative EU/EEA/CH member state (alongside Germany) to document the treatment without coining a bloc key. Same duration rule and same non-qualification as the DE observation: 30-day per-entry stay, VISIT only, no STUDY/WORK_REMOTE/long RESIDE; a longer stay requires a visa category. See the DE observation's note for the full verbatim enumerated EU/EEA/CH list and the duration quote. Keyed "FR" per PASSPORT.md (member states enumerated individually; no bloc coined). effectiveAt 2019-02-01 from secondary reporting; the MFA page states no date. | seed | seed | 572bb983-8d95-4cd7-a6ac-0d85844b0587 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 138 | b63ecadc-1176-4e81-92ad-bf3aa7022ac5 | uz | visa-access | {"route":"Electronic entry visa (e-visa.gov.uz) — tourist/short stay","permits":["VISIT"],"maxStayDays":30,"passportSet":"IN"} | NULL | https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi | Ministry of Foreign Affairs of the Republic of Uzbekistan | 2018-07-15 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SHORT-STAY VISIT ONLY — DOES NOT QUALIFY. India (ordinary passport) is served by the electronic entry visa: "Republic of India" is item 31 in Appendix 1, the MFA's "List of countries whose citizens have the opportunity to obtain an electronic entry visa (e-visa.gov.uz)". India is NOT on the main 30-day visa-free list. The e-visa fee schedule on the same page, verbatim: "For a single-entry electronic visa - 20 USD; For a double-entry electronic visa for up to 30 days - 35 USD; For a multiple-entry electronic visa for up to 30 days - 50 USD" — so the e-visa per-entry stay ceiling is 30 days. IMPORTANT NATIONALITY NUANCE, verbatim from Note 6: Uzbekistan has established a 60-day visa-free regime "for holders of DIPLOMATIC passports" of, among others, India — and Note 6 explicitly carves India out of the accreditation clause ("except India"). That 60-day figure therefore applies to DIPLOMATIC passports only, not ordinary Indian passports, and I do not extend it to the scenario cohort. India also appears in Appendix 2 (5-day visa-free transit). None of these grant STUDY, WORK_REMOTE, or RESIDE beyond a short visit. effectiveAt 2018-07-15 is the widely reported start of Uzbekistan's e-visa system; the page states no date (UNVERIFIED against the primary decree). | seed | seed | f0b3f13c-8f7a-42f9-b512-a1f8e7f23bb2 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 139 | b5108a76-b2eb-489a-b28a-3f9a6eccaafa | uz | visa-access | {"route":"Visa-free entry (10-day) with onward ticket","permits":["VISIT"],"maxStayDays":10,"passportSet":"CN"} | NULL | https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi | Ministry of Foreign Affairs of the Republic of Uzbekistan | 2024-01-01 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SHORT-STAY VISIT ONLY — DOES NOT QUALIFY. China (ordinary passport) short-stay treatment, verbatim from the MFA page. Item 90 under "Up to 10 days:" reads "The People's Republic of China, including the Special Administrative Regions of the People's Republic of China, Hong Kong and Macao." Note 4, verbatim: "Citizens of Bahrain, Qatar, Kuwait, Oman and China are granted a visa-free regime for a period of 10 days, provided they have a flight ticket to their home country or a third country." maxStayDays 10 is that per-entry visa-free ceiling. OTHER CHINA SHORT-STAY OPTIONS on the same page, all still VISIT and all short: (a) Note 5 grants 30-day visa-free tourism to "citizens of China (as part of tourist groups)... who have reached the age of 55"; (b) China ("People's Republic of China (including Hong Kong)") is item 11 in Appendix 1, the e-visa list, so a 30-day e-visa is available. None grants STUDY, WORK_REMOTE, or RESIDE beyond a short visit. effectiveAt 2024-01-01 is approximate (the 10-day China visa-free regime was reported from 2024); the page states no date (UNVERIFIED against the primary decree). | seed | seed | e05915b2-e07f-46eb-9e0b-c4fee141a3d0 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 140 | f71d0e66-ffc2-4cca-b19b-e9fd29588c5b | uz | visa-access | {"route":"IT Visa (IT Park Uzbekistan) — multi-entry, up to 3 years","permits":["RESIDE","STUDY"],"maxStayDays":1095,"passportSet":"*"} | NULL | https://it-park.uz/en/itpark/news/how-to-become-a-member-of-it-park-uzbekistan | IT Park Uzbekistan (state technopark agency) | 2022-03-01 | 2026-07-19T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | THE CLOSEST ROUTE TO QUALIFYING, AND THE KEY NEAR-MISS — but I do NOT record it as granting WORK_REMOTE, and that is the whole finding. SOURCE QUOTE, verbatim from IT Park's own program page: "IT Visa. Foreign specialists, investors, and founders of IT Park Uzbekistan members are granted a multi-entry visa valid for up to 3 years. The IT Visa allows foreigners to: access medical care and educational services equally with Uzbek citizens; obtain temporary registration for the visa duration; and purchase real estate of any value." Corroborated verbatim by IT Park's founder-experience page (https://www.it-park.uz/en/itpark/news/how-i-got-my-it-card-and-it-visa-in-10-days-a-founder-s-experience-in-uzbekistan): "The IT Visa is a multiple-entry visa valid for up to three years. It allows you to legally live and work in Uzbekistan without needing a separate work permit" and the IT Card gives the right to "use medical and educational services in public institutions at the same rates as Uzbek citizens... obtain residence registration in any region of Uzbekistan for the validity period of the IT Card." RESIDE: GRANTED — "obtain temporary registration for the visa duration" / "residence registration... for the validity period". This is continuous residence for the visa's whole term, not a resetting per-entry tourist cap, so maxStayDays 1095 (3 years) is the sourced RESIDE ceiling; no shorter per-entry stay is published for this residence-type visa. STUDY: recorded because the source affirmatively grants "access [to] educational services equally with Uzbek citizens" — a study-access right. SCOPE CAVEAT (why this is weaker than it looks): the founder page scopes it to "educational services IN PUBLIC INSTITUTIONS at the same rates as Uzbek citizens"; enrolment at a private/foreign school (which is what the scenario's STUDY at a Network School node would be) is NOT addressed, so STUDY at a private foreign institution is UNKNOWN, not affirmatively granted. WORK_REMOTE: NOT RECORDED — the deciding point the tasking flagged. Every primary/official description ties the IT Visa's work right to IT PARK RESIDENT STATUS: it is granted only to "foreign specialists, investors, and founders of IT Park Uzbekistan members", and an eligible "IT specialist" is, per the government services portal (https://oldmy.gov.uz/en/service/1244), verbatim, "a specialist who has a qualification/specialization in the field of information technology, IS EMPLOYED IN A LEGAL ENTITY OF A RESIDENT of the Republic of Uzbekistan in an IT specialty and confirms his income from IT activities is not less than the equivalent of 30,000 US dollars for the last 12 months." Founders/investors must likewise be tied to an Uzbekistan-registered IT Park resident legal entity ($10,000 investment). The work the IT Visa authorises is therefore work FOR/THROUGH an Uzbek IT Park resident company (no separate work permit needed) — NOT remote work for a foreign employer, which is the scenario's WORK_REMOTE. I found no primary statement that a person working remotely for a foreign employer, unconnected to an IT Park resident, may hold this visa. So the IT Visa is RESIDE + (public-institution) STUDY for up to 3 years, but it is a near-miss on the scenario's WORK_REMOTE. PASSPORT-SET BASIS FOR "*": the WEAKER form, stated explicitly per PASSPORT.md §1 — the eligibility conditions are professional/income/investment tests (IT qualification, $30,000 income, or $10,000 investment) with NO nationality condition stated anywhere on the IT Park pages or the government services portal, and I did NOT find an affirmative government statement that the IT Visa is open to all nationalities. TIER: it-park.uz is the state technopark agency describing its own program (OFFICIAL_SECONDARY); the visa is issued via the MFA/Migration department on IT Park's recommendation. I could NOT retrieve the primary Presidential Decree / Cabinet Resolution establishing the IT Visa from lex.uz in this pass, so effectiveAt 2022-03-01 (IT Visa introduced in early 2022) is approximate and UNVERIFIED against the primary instrument. | seed | seed | a37907d2-b9c8-4611-82b2-9b0efb45aab7 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 141 | 6421b1ba-a2bf-48fe-89ca-e787d29fee52 | uz | visa-access | {"route":"INV — Multiple-entry investment visa (up to 3 years)","permits":["RESIDE"],"maxStayDays":1095,"passportSet":"*"} | NULL | https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi | Ministry of Foreign Affairs of the Republic of Uzbekistan | 2024-01-01 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | A LONG-STAY RESIDE ROUTE, recorded as a near-miss because it clears the 180-day duration but is silent on STUDY and WORK_REMOTE. SOURCE QUOTE, verbatim, category 28 on the MFA visa page: "INV — Multiple-entry investment visa. It is issued upon the application of enterprises investing in the economy of the Republic of Uzbekistan in an amount not less than 8500 times the base payment established in the Republic of Uzbekistan when making foreign investments, for the purpose of acquiring shares and stakes in economic associations, as well as for the establishment of enterprises with the participation of foreign investments, with the possibility of extending its term without leaving the territory of the Republic of Uzbekistan, issued to foreign citizens and stateless persons, valid for up to 3 years. In this case, the investor's family members (spouse, parents, and children) are issued a visitor visa for the duration of the 'Investment visa' with the right to extend it without leaving the republic." RESIDE: granted for up to 3 years (continuous, extendable without leaving), so maxStayDays 1095. STUDY: NOT stated — silence is not permission, recorded as unknown, not granted. WORK_REMOTE: NOT stated — the visa's purpose is investment (acquiring shares / establishing enterprises), not remote work for a foreign employer; no work authorisation for a foreign employer is granted. Also note the high threshold (8500x the base payment). PASSPORT-SET BASIS FOR "*": weaker form — the condition is an investment test, no nationality condition stated ("issued to foreign citizens and stateless persons"), and no affirmative universality statement found. maxStayDays 1095 is the visa validity/residence term; the page states no shorter per-entry cap for this residence-type investment visa. | seed | seed | 447e575c-ffdf-4a54-bd1a-99f45b4b3c91 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 142 | c1584ec1-14f4-46fb-bcae-83f7a3feefd7 | my-jhr-forest-city-b1 | capacity | 245 | beds | NULL | FIXTURE | 2026-07-17 | 2026-07-17T00:00:00.000Z | FIXTURE | UNVERIFIED | ANALYST_ENTRY | NULL | FIXTURE. No property record has been sourced for any specific Forest City building — no title, no listing, no capacity document. The 245 is invented and must never resolve a gate or move a score. WHY THIS STILL EXISTS, 2026-07-17. The original note said 'replace it with a sourced record or delete it', and it was written when this was the only property in the repo. It has since acquired a job that outlived that instruction: it is the seeded-data proof that a FIXTURE observation is DISCOUNTED rather than used. tests/unit/forest-city.test.ts asserts it by name — 'never lets the FIXTURE building resolve a gate or move the score' — checking the capacity gate reads NEEDS_EVIDENCE with FIXTURE named in the reason, and that beds-headroom is unusable with a null value. scoring.config.json gives the FIXTURE tier weight 0.0, so this is proof and not decoration. tests/db/checks.sql additionally uses this entity for its chain and cycle fixtures. The 'exercise the property level' job it was created for IS now dead — my-jhr-forest-city-golf-hotel is a real sourced PROPERTY under the same parent and does that with real data. But deleting this would cost the two guards above to remove a number that is provably inert. If a sourced record for a specific Forest City building ever lands, it belongs in its own candidate file beside the golf hotel; this entity stays as the FIXTURE guard until something else proves the same property on seeded data. | seed | seed | c128c365-2314-41eb-bb2b-2630c9aff015 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 143 | d42226f8-01ca-4e7f-b47e-4b2655f118fd | uz | visa-access | {"route":"A-1 — Student entry visa (for the period of study)","permits":["RESIDE","STUDY"],"maxStayDays":365,"passportSet":"*"} | NULL | https://gov.uz/en/mfa/activity_page/o-zbekiston-respublikasi-vizasi | Ministry of Foreign Affairs of the Republic of Uzbekistan | 2024-01-01 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | A RESIDE + STUDY ROUTE, recorded as a near-miss because it clears both duration (up to 1 year) and STUDY but grants NO work of any kind (and certainly not WORK_REMOTE). SOURCE QUOTE, verbatim, category 20 on the MFA visa page: "A-1 — Student entry visa (permanent - for the period of study). It is issued for a period of up to 1 year to foreign students studying at educational institutions located in the territory of the Republic of Uzbekistan, and is issued upon requests from educational institutions, ministries, departments and organizations of the Republic of Uzbekistan that have an educational institution under their jurisdiction." RESIDE: granted for up to 1 year → maxStayDays 365. STUDY: granted, but BOUND to enrolment at "educational institutions located in the territory of the Republic of Uzbekistan" and requires the sponsoring institution to request the visa — so it presupposes an Uzbekistan-recognised educational institution as sponsor; whether a Network School node could sponsor it is UNKNOWN (an operating-licence question, not answered here). WORK_REMOTE: NOT granted — no work authorisation attaches to a student visa. (The related category 19 'STD' is the short-term exchange student visa, also up to 1 year, same non-work character.) PASSPORT-SET BASIS FOR "*": weaker form — the category is purpose-based (study) with no nationality condition stated; no affirmative universality statement found. maxStayDays 365 is the issue/stay period stated by the category. | seed | seed | 420e71af-2615-4e22-abda-7ebe289c8a7d | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 144 | e2979f54-93ff-4c16-a409-f16fb0485ca4 | uz | operating-licence | true | NULL | https://lex.uz/en/docs/-5013007 | Law of the Republic of Uzbekistan "On Education" No. O‘RQ-637 of 23 September 2020 (consolidated text, National Database of Legislation, lex.uz) | 2020-09-24 | 2026-07-19T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | DOCUMENT_EXTRACTION | NULL | A FOREIGN entity MAY hold the licence/authorisation to establish and run an educational institution in Uzbekistan. This is the opposite of Thailand's Private School Act (which required Thai nationality of a natural-person licensee and Thai-majority ownership of a juristic-person licensee): the Uzbek Law 'On Education' (O‘RQ-637, in force 24.09.2020; consolidated text reflects amendments through 2026) imposes NO nationality or ownership condition on the licence holder, and it expressly enables branches/campuses of FOREIGN higher-education organizations. Chain of reasoning, all from the quoted consolidated text on lex.uz: (A) Art. 3 defines a 'non-state educational organization' (nodavlat taʼlim tashkiloti) simply as a 'legal entity' (yuridik shaxs) that provides educational services on the basis of a licence or in notification procedure — the definition contains no citizenship or share-ownership qualifier. (B) Art. 31 makes the licence the general operating requirement for non-state educational organizations, again with no nationality condition on who the legal entity may be. (C) Art. 29 states that educational organizations of Uzbekistan and their branches, AND branches of educational organizations of FOREIGN states, are established with legal-entity status — i.e. foreign educational organizations' branches are contemplated as licensable legal persons; the third paragraph confirms an organization is entitled to conduct educational activity from the date the licence is obtained. (D) Art. 30 is the express foreign-branch enabling provision: branches (centers, educational campuses) and other subdivisions of foreign higher-education organizations MAY be established on the territory of the country, are established as legal entities in Uzbekistan, and their operating conditions are fixed by resolutions of the President or the Cabinet of Ministers under Uzbekistan's international treaties. I searched the licensing chapter (Chapter 4, Arts. 29-34, and the state-body powers in Art. 23/43) for any citizenship-of-founder requirement, majority-local-ownership requirement, or express foreign prohibition (Uzbek search terms: 'fuqarosi boʻlishi' = must be a citizen; 'xorijiy ... taqiq'/'man etiladi' = foreign ... prohibited) and found NONE — there is no nationality/ownership bar on the licence holder anywhere in the statute. SCOPE: the EXPRESS foreign enabling in Art. 30 is at HIGHER-education level (xorijiy oliy taʼlim tashkilotlari), which is why ~30 foreign university branches operate in Tashkent (Westminster/WIUT, Inha, Webster, MDIS, Turin, etc.); at other levels no express foreign-branch clause exists, but the general non-state-organization licensing (Arts. 3, 31) is equally free of any nationality condition, so a foreign-controlled legal person is not barred there either. CONDITION, not a bar: Art. 30's foreign HE-branch route runs 'in accordance with international treaties' via Presidential/Cabinet resolution (the intergovernmental mechanism behind WIUT/Inha/etc.); that is a modality of authorisation, not a nationality exclusion — the licence/authorisation remains available to the foreign entity. CIRCUMSTANTIAL CORROBORATION (secondary, not the primary proof): government and press sources state ~30 branches of foreign educational institutions operate in Uzbekistan, and Presidential Resolution PP-2909 of 20.04.2017 sits in the enabling regime; these support but do not substitute for the statute above. FETCH FACTS (not missing legal facts, per D-043): the ILO NATLEX official English PDF of O‘RQ-637 (natlex.ilo.org/.../116503/UZB-116503.pdf) and the Presidential decree PDF on invest.miit.uz both triggered a browser DOWNLOAD the fetcher could not render (outcome NAVIGATION_FAILED, 'Download is starting') — reachable in a real download session, just not renderable here; cis-legislation.com serves an AI English translation but paywalls after Art. 8. The lex.uz consolidated text (this sourceUrl) was fully retrieved and is authoritative but rendered in Uzbek Latin script despite the /en/ path; the verbatim quote below is that Uzbek text, with a working English translation, cross-checked against the cis-legislation English rendering of the same definitions. Recorded TRUE because the licensing regime was actually located and it is affirmatively open to foreign entities (with an express foreign-branch enabling article); the value is the statute's answer, not a legal opinion. SOURCE QUOTE, verbatim from lex.uz (O‘RQ-637 consolidated), Uzbek Latin script, with translation: Art. 29 (Taʼlim tashkilotlarining huquqiy maqomi): «Oʻzbekiston Respublikasining taʼlim tashkilotlari va ularning filiallari, shuningdek xorijiy davlatlar taʼlim tashkilotlarining filiallari yuridik shaxs maqomi bilan tashkil etiladi.» — 'Educational organizations of the Republic of Uzbekistan and their branches, as well as branches of educational organizations of foreign states, shall be established with the status of a legal entity.' Art. 30 (Taʼlim tashkilotlarini tashkil etish, qayta tashkil etish va tugatish): «... mamlakat hududida xorijiy oliy taʼlim tashkilotlarining filiallari (markazlari, taʼlim kampuslari) va boshqa boʻlinmalari tashkil etilishi mumkin. Xorijiy oliy taʼlim tashkilotlarining boʻlinmalari Oʻzbekiston Respublikasida yuridik shaxs maqomi bilan tashkil etiladi. Xorijiy oliy taʼlim tashkilotlarining, shuningdek xalqaro tashkilotlarning taʼlim muassasalari filiallari (markazlari, taʼlim kampuslari) va boshqa boʻlinmalari faoliyatini tashkil etish masalalari boʻyicha Oʻzbekiston Respublikasining xalqaro shartnomalariga muvofiq Oʻzbekiston Respublikasi Prezidentining yoki Vazirlar Mahkamasining qarorlari qabul qilinadi.» — '... on the territory of the country, branches (centers, educational campuses) and other subdivisions of foreign higher-education organizations may be established. Subdivisions of foreign higher-education organizations shall be established in the Republic of Uzbekistan with the status of a legal entity. On the questions of organizing the activity of branches (centers, educational campuses) and other subdivisions of foreign higher-education organizations, as well as of educational institutions of international organizations, resolutions of the President of the Republic of Uzbekistan or of the Cabinet of Ministers shall be adopted in accordance with the international treaties of the Republic of Uzbekistan.' Art. 31 (Nodavlat taʼlim tashkilotlari), first part (as amended by O‘RQ-1091 of 23.10.2025): «Nodavlat taʼlim tashkilotlari oʻz faoliyatini litsenziya asosida amalga oshiradi, bundan qoʻshimcha taʼlimni, kadrlarni qayta tayyorlash va ularning malakasini oshirishni amalga oshiruvchi taʼlim tashkilotlari mustasno.» — 'Non-state educational organizations carry out their activity on the basis of a licence, except for educational organizations providing additional education, retraining and advanced training of personnel.' Art. 3 definition (nodavlat taʼlim tashkiloti): «... litsenziya asosida yoki xabardor qilish tartibida taʼlim xizmatlari koʻrsatuvchi yuridik shaxs» — 'a legal entity providing educational services on the basis of a licence or under the notification procedure' (no nationality or ownership qualifier). | seed | seed | 8b2247b5-dd9e-47e1-bc70-0e1a88e79b95 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 145 | a5cecd44-6436-4711-8b54-a129b7b3c5dc | vn | visa-access | UNRESOLVED | NULL | https://hanoi-be.mfa.gov.tr/Content/assets/consulate/images/localCache/1/354983aa-6b1a-4df5-b559-e94969bfc05c.pdf | National Assembly of Vietnam, Law No. 47/2014/QH13 (copy hosted by the Turkish Embassy in Hanoi) | 2026-07-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed {"US": [{"route": "DH visa (study/internship) \u2014 Art. 8(12), Law 47/2014/QH13", "permits": ["RESIDE", "STUDY"], "maxStayDays": 365}, {"route": "U\u01101 vis and an independent agent re-fetched the source and rejected it. VERIFIER: PDF loaded (WebFetch could not parse the FlateDecode streams, so I extracted it locally with pdftotext — 734 lines). Every quoted fragment is verbatim: Art. 8(12) DH, 8(16) LD, 8(17) DL, 8(19) VR; Art. 9(2) 03 months, 9(3) 06 months, 9(4) 12 months, 9(5) 02 years, 9(6) 05 years; Art. 10(4)(c) and (d). But the document is the ORIGINAL, UNAMENDED Law No. 47/2014/QH13 dated June 16, 2014 (header line 2). Its Article 8 enumerates only 20 symbols and STOPS AT ITEM 20 (SQ) — there is no item 22 and no UĐ1. Case-insensitive greps return ZERO hits for each of: UĐ1, UD1, ĐT1, LĐ2, DN1, e-visa, electronic, SVEC, '45 day', '90 day'. The 2014 text has only the undifferentiated symbols DT, DN, LĐ — not the numbered ĐT1/LĐ2/DN1 variants (those come from Law 51/2019/QH14, a different instrument not in this PDF). So of ~9 routes claimed per country, this source supports only three: DH=365d (Art. 9(4) 12 months), DL=90d (Art. 9(2) 03 months), VR=180d (Art. 9(3) 06 months), plus DH permits STUDY via Art. 10(4)(d). UNSUPPORTED by this source: UĐ1 1825d (symbol absent entirely; also cited to Law 118/2025 which this PDF predates by 11 years), ĐT1 1825d, LĐ2 730d, DN1 365d, E-visa 90d, SVEC 90d, and the EU 45-day unilateral exemption. Additionally, nothing in this document differentiates US/EU/IN/CN — the matrix asserts four nationality-specific route sets that this source cannot ground. The 'VND 100bn' ĐT1 capital threshold appears nowhere. No WORK_REMOTE was claimed, correctly. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: FAIL, on positive evidence. Vietnam's visa system is a CLOSED, PURPOSE-BOUND ENUMERATION: Art. 8 lists every symbol that exists and Art. 9 caps each one's duration. No symbol permits RESIDE+STUDY+WORK_REMOTE together. NEAREST MISS: DH permits RESIDE and STUDY for up to 365 days but is silent on remote work — silence is not permission. Every other long route (UĐ1 1825d, ĐT1 1825d, LĐ2 730d) permits RESIDE but is purpose-bound to a Vietnamese host and is silent on both STUDY and remote work for a FOREIGN employer. VR is exactly 180 days but is RESIDE-only. Art. 10(4) confirms purposes are not fungible: study requires 'written acceptance of a school', work requires a 'work permit' — i.e. the route must match the activity, so no single symbol covers all three. I did NOT assume the expected failure: I chased two instruments post-dating the base statute. (1) Decree 221/2025/NĐ-CP (SVEC, in force 2025-08-15) — 5-yr card but only 90 days' temporary residence per entry, so it fails duration. (2) Law 118/2025/QH15 created NEW symbols UĐ1/UĐ2, in force 2026-07-01 — SIXTEEN DAYS AGO. UĐ1 is 5 years and is aimed squarely at digital-tech talent, so it was the strongest candidate; it still fails because it is granted to 'nhân lực công nghiệp công nghệ số chất lượng cao' (high-quality personnel OF Vietnam's digital technology industry), requires a Vietnamese nominating institution, still requires a work permit or exemption to work, and says nothing about study or about remote work for a foreign employer. Vietnam has NO digital-nomad visa. Matrix is nationality-neutral — Art. 8 draws no distinction by passport, so US/EU/IN/CN receive the same symbols; the only cohort difference is the 45-day unilateral exemption, which reaches only 7 EU states (DE, FR, IT, ES, DK, SE, FI) and is RESIDE-only anyway. Quote is Art. 8/9 verbatim via pdftotext. Tiered OFFICIAL_SECONDARY, not PRIMARY: this is a foreign ministry's hosted copy of the Vietnamese law. Vietnam's own database vbpl.vn timed out 3x — see blocked. QUOTE THE RESEARCHER OFFERED: “Article 8. Visa's symbols [...] 12. DH - Issued to people who come to study or serve internship. [...] 16. LD - Issued to people who come to work. 17. DL - Issued to tourists. [...] 19. VR - Issued to people who come to visit their relatives or for have other purposes. [...] Article 9. Visa's duration [...] 2. The duration of a HN or DL visa is no longer than 03 months. 3. The duration of a VR visa is no longer than 06 months. 4. The duration of a NG1, NG2, NG3, NG4, LV1, LV2, DN, NN1, NN2, NN3, DH, PV1, PV2 or TT visa is no longer than 12 months. 5. The duration of a LĐ visa is no longer than 02 years. 6. The duration of a ĐT visa is no longer than 05 years. [...] Article 10. Conditions for issuing visa [...] 4. Proof of entry purposes must be provided when applying for a visa in the following cases: [...] c) Any foreigner who comes to work in Vietnam must have work permits in accordance with the Labor Code; d) Any foreigner who comes to study in Vietnam must have written acceptance of a school or an education institution in Vietnam.” | seed | seed | b7e7ef14-a652-4caf-a888-9079dd01a90b | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 146 | d249d77b-95e9-4890-94e7-c55772e3e990 | vn | visa-access | UNRESOLVED | NULL | https://hvannd.edu.vn/mot-so-diem-moi-cua-luat-so-1182025qh15-sua-doi-bo-sung-mot-so-dieu-cua-10-luat-co-lien-quan-den-an-ninh-trat-tu | Học viện An ninh nhân dân (People's Security Academy, Ministry of Public Security) | 2026-07-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | SUPPORTING EVIDENCE for the UĐ1/UĐ2 rows in the matrix above — filed separately because it is the sole source establishing the newest route, which entered force 2026-07-01 (16 days before retrieval) and is therefore the single most likely thing for a stale database to miss. Value is UNRESOLVED because this row carries no independent gate value; the matrix row above is the claim. The quote establishes (a) UĐ1 is granted to high-quality DIGITAL TECHNOLOGY INDUSTRY PERSONNEL — a Vietnam-employment purpose, not a remote-work-for-foreign-employer purpose; (b) its duration is capped at 5 years; (c) a matching temporary residence card runs to 10 years. Nothing in the text permits study, and nothing permits remote work for a foreign employer. Published by an academy of the Ministry of Public Security — the ministry that IS the immigration authority — hence OFFICIAL_SECONDARY. I could not reach the National Assembly's own consolidated text of Law 118/2025/QH15; bocongan.gov.vn carried only the Điều 6a data-reuse amendment, not the Art. 8/9 text. SOURCE QUOTE, verbatim from the page on the retrieval date: “Thị thực ký hiệu UĐ1: cấp cho người nước ngoài là nhân lực công nghiệp công nghệ số chất lượng cao [...] Thị thực ký hiệu UĐ2: cấp cho vợ, chồng, con dưới 18 tuổi của người nước ngoài được cấp thị thực UĐ1 [...] quy định thị thực ký hiệu LS, ĐT1, ĐT2, UĐ1, UĐ2 có thời hạn không quá 05 năm [...] quy định thẻ tạm trú ký hiệu ĐT1, UĐ1, UĐ2 có thời hạn không quá 10 năm” | seed | seed | 135dbe1f-eca4-4bf4-8c7d-28c1453764b9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 147 | 861d7a74-e2ab-449f-a19a-23588e855b3d | vn | visa-access | UNRESOLVED | NULL | https://xaydungchinhsach.chinhphu.vn/nghi-dinh-so-221-2025-nd-cp-mien-thi-thuc-co-thoi-han-cho-nguoi-nuoc-ngoai-thuoc-dien-doi-tuong-dac-biet-119250809110419124.htm | Cổng Thông tin điện tử Chính phủ (Government of Vietnam policy portal) | 2025-08-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | SUPPORTING EVIDENCE for the SVEC row in the matrix. Value UNRESOLVED — the matrix row above is the claim. This kills the 'Vietnam golden visa' story that relocation media report as a long-stay route: Decree 221/2025/NĐ-CP grants a card valid up to 5 years, BUT the quote is explicit that each entry confers only 90 DAYS of temporary residence. 90 < 180, so it fails duration on one grant regardless of purpose. Extension beyond 90 days is expressly discretionary, so per the gate's rule it must not be counted toward maxStayDays. The decree also confers no right to work or study — it is an ENTRY exemption only. Eligibility is institutional (scholars, investors, corporate leaders, digital-tech personnel) and requires Vietnamese nomination. Note the separate, widely-reported '10-year golden visa' for investors is NOT enacted: no instrument, no application process — that is marketing, and per rule 9 I did not infer it upward into a route. SOURCE QUOTE, verbatim from the page on the retrieval date: “Người nước ngoài sử dụng thẻ miễn thị thực đặc biệt được cấp chứng nhận tạm trú 90 ngày cho mỗi lần nhập cảnh. [...] Thời hạn miễn thị thực không quá 05 năm và ngắn hơn thời hạn còn lại của hộ chiếu ít nhất 30 ngày.” | seed | seed | 13aa2426-bd77-41d7-9aa9-96fa8afda7b6 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 148 | 84cb21ee-1a6e-412c-95b6-22c99afeef1a | vn | visa-access | UNRESOLVED | NULL | https://evisa.xuatnhapcanh.gov.vn/trang-chu-ttdt | Cục Quản lý xuất nhập cảnh (Vietnam Immigration Department) | 2023-08-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | SUPPORTING EVIDENCE for the E-visa row. Value UNRESOLVED — the matrix row above is the claim. Straight from the immigration authority's own national e-visa portal: the e-visa caps at 90 days, so it fails the 180-day bar before purpose is even reached. Recorded because the e-visa is the route ~all four cohorts would actually reach for (available to all countries/territories since 2023-08-15), and because it is the route most often mis-sold by relocation blogs as a nomad-compatible option. It is not: it is an entry permission with no work or study right attached. SOURCE QUOTE, verbatim from the page on the retrieval date: “Vietnam E-visa is valid for maximum of 90 days, single or multiple entry.” | seed | seed | 2db114fa-3740-4544-af6b-3f44a6d8765e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 149 | 0d62c8dc-d92f-480d-b066-16c26470219f | my-jhr-forest-city-b1 | beds-headroom | 95 | beds | NULL | FIXTURE | 2026-07-17 | 2026-07-17T00:00:00.000Z | FIXTURE | UNVERIFIED | ANALYST_ENTRY | NULL | FIXTURE. Beds above the scenario's 150 floor, derived from the invented 245 in the capacity note. Carries zero confidence by construction. | seed | seed | 17f2daad-c99c-412f-a0cd-cb70957bc15c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 150 | aba685bb-f9e2-4e3e-91ff-8fa9eeb4579f | my-jhr-forest-city-golf-hotel | capacity | {"rooms":298} | rooms | https://fcgolfhotel.com/about-forest-city-phoenix-golf-hotel-johor-bahru | Country Garden Pacificview Sdn. Bhd. (operator) | 2026-07-17 | 2026-07-17T00:00:00.000Z | SELF_REPORTED | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | 298 ROOMS IS MEASURED. THE BED COUNT IS NOT, AND IS NOT DERIVABLE (D-040). SOURCE QUOTE, verified verbatim by an adversarial verifier that re-fetched this page (D-032): 'The hotel provides 298 spacious rooms with a variety of room types including King and Twin Guest Rooms, along with 3 Presidential Suites and a Family Suite.' The verifier also read the page's own per-type table — Superior Room 168, Deluxe Room 87, Superior Suite 22, Deluxe Suite 15, Disabled Rooms 2, Family Suite 1, Presidential Suite 3 — which sums to exactly 298 and is stated on the page as 'Total Room = 298'. WHY UNVERIFIED AND NOT CORROBORATED. A live Trip.com property record independently states 'Number of Rooms: 298' (plus 'Local hotel name: Forest City Golf Hotel', 'Opened: 2018'), read in a browser on 2026-07-17. THE VERIFIER COULD NOT READ IT: trip.com returns HTTP 418 to a fetcher and bot-blocks it. Blocked is not dead, but a claim no second party can re-check has not been corroborated — so this row rests on the operator's page alone and carries UNVERIFIED confidence, which is pass-eligible but weak by design. Do not restore CORROBORATED until a checker can read the second source. WHY THIS IS A {rooms} CLAIM AND NOT A BED COUNT. No source publishes beds. The verifier attempted to refute this across the operator site, Booking.com, Klook, Hotels.com, Expedia, Agoda and TripAdvisor and could not — and found a STRONGER basis than the research had: the operator SELLS 'Superior/Deluxe King Room' and 'Superior/Deluxe Twin Room' as separate products, but COUNTS only 'Superior Room 168' and 'Deluxe Room 87', collapsing King and Twin into one number. So 255 of 298 rooms have an UNKNOWN King/Twin split, spanning 1-bed and 2-bed configurations, and beds are bounded only to roughly 298–553 — non-derivable from the operator's own data, without reference to any blocked page. Beds per room is therefore a policy about how we intend to house members, not a fact about this building; it lives on the scenario (bedsPerRoom) and the gate names the conversion in its reason. At the shipped policy of 1.0 this yields 298 beds — a FLOOR, since a guest room holds at least one bed — which clears the 150 requirement without over-claiming. EFFECTIVE DATE IS THE WEAKEST PART OF THIS ROW. This page is UNDATED; its only temporal marker is '© 2024 COUNTRY GARDEN PACIFICVIEW SDN. BHD (1044344-D)'. effectiveAt is set to the retrieval date because a live sales page asserts present inventory, and because a live OTA record read the same day states the same 298 — but if the operator's page is a 2024 artefact left untouched, this figure could be ~2 years stale and would age past the 365-day window. Forest City has been subject to well-publicised occupancy and viability pressure that nobody has investigated and that could have closed or repurposed inventory. A dated source, or an operator confirmation, would settle it. | seed | seed | b95dc7fc-2cad-4970-9d85-e30c682eb4cb | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 151 | a211aeb3-7db8-44dc-a703-67bdbd27f34c | vn | visa-access | UNRESOLVED | NULL | https://vietnamembassy-usa.org/news/2023/10/new-policy-electronic-visa-and-visa-exemption | Embassy of the Socialist Republic of Vietnam in the United States | 2023-08-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | NEEDS_EVIDENCE | DOCUMENT_EXTRACTION | NULL | SUPPORTING EVIDENCE for the EU cohort's 45-day unilateral-exemption row, and corroboration of the 90-day e-visa cap from a second official Vietnamese body. Value UNRESOLVED — the matrix row above is the claim. Cohort mapping of the quoted list: of the 13 exempt states, only 7 are EU members (Germany, France, Italy, Spain, Denmark, Sweden, Finland) — the UK, Norway, Russia, Japan, Korea and Belarus are not, so this does NOT cover the EU cohort uniformly and I did not model it as such. NO exemption exists for the US, IN or CN cohorts. The exemption is 45 days and is an entry exemption only — it confers no work or study right — so it fails on both duration and purpose. This is the row that would have tripped the tourist-duration bug: it is short AND purposeless, and neither fact rescues the other. SOURCE QUOTE, verbatim from the page on the retrieval date: “maximum of 90 days with single or multiple entry (ies) [...] Federal Republic of Germany, Republic of France, Republic of Italy, Kingdom of Spain, United Kingdom of Great Britain and Northern Ireland, Russian Federation, Japan, Republic of Korea, Kingdom of Denmark, Kingdom of Sweden, Kingdom of Norway, Republic of Finland and Republic of Belarus” | seed | seed | 5df121e6-192f-438f-a0e4-75e230ea2437 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 152 | 66082e7d-4da9-4e9b-90e0-e7c915f5855b | vn | visa-access | {"route":"E-visa (thị thực điện tử, symbol EV) — Art. 8(21) Law on Entry/Exit as amended by Law 23/2023/QH15","permits":["VISIT"],"maxStayDays":90,"passportSet":"*"} | NULL | https://evisa.xuatnhapcanh.gov.vn/trang-chu-ttdt | Cục Quản lý xuất nhập cảnh (Vietnam Immigration Department) | 2023-08-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE, verbatim from the Immigration Department's own e-visa portal on the retrieval date: "An electronic visa (E-visa) is one of visa types issued to foreigners by Vietnamese Immigration Department via electronic system. Vietnam E-visa is valid for maximum of 90 days, single or multiple entry." Conditions stated: "Outside Vietnam foreigners; Holding valid passport; Not falling into the cases of suspension from entry prescribed in Article 21 of Law on foreigners' entry into, exit from, transit through, residence in Vietnam." WHAT IT PERMITS: entry and temporary stay only. It grants no right to work, no right to study, and no residence status — BRIEF.md §6's exact case: tourist-class access is not permission for the activity. WHAT IT FAILS FOR THE SCENARIO: 90 < 180, so it fails the duration bar before purpose is even reached; it also permits none of RESIDE/STUDY/WORK_REMOTE. PASSPORT-SET BASIS FOR "*": this page states an eligibility test (outside Vietnam, valid passport, not Art. 21-barred) with no nationality condition; the affirmative nationality-neutrality statement (Resolution 127/NQ-CP: citizens of ALL countries and territories) is on a separate official page and is filed as its own observation below. The portal notes new application domains since 2024-11-11 (thithucdientu.gov.vn, evisa.gov.vn); this page remains the Department's live informational portal. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, https://evisa.xuatnhapcanh.gov.vn/trang-chu-ttdt fetched 2026-07-18. Quote verbatim as claimed. Page grants entry/stay only — no work/study/residence right stated, so permits [VISIT] is right. Eligibility conditions state no nationality condition, matching the claim's disclosed (weaker, absence-based) basis for '*'; the affirmative all-nationalities statement is correctly filed as claim 1. Art. 31 of the consolidated law ('Thời hạn tạm trú cấp bằng thời hạn thị thực') supports 90 as a single-route per-entry stay for EV — no DL-style 30-day carve-out applies to EV. | seed | seed | 3c42d7c2-50c9-416f-8e8b-de35e7c94916 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 153 | db65198d-7057-4b4d-854e-3d2f7d0df7fd | vn | visa-access | {"route":"E-visa (EV) — Resolution No. 127/NQ-CP (universal eligibility)","permits":["VISIT"],"maxStayDays":90,"passportSet":"*"} | NULL | https://vietnamembassy-usa.org/news/2023/10/new-policy-electronic-visa-and-visa-exemption | Embassy of the Socialist Republic of Vietnam in the United States | 2023-08-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | Second observation of the same EV route, filed separately because this page carries the AFFIRMATIVE nationality-neutrality statement the portal lacks. SOURCE QUOTE, verbatim: "On August 14, 2023, the Government of Viet Nam issued Resolution No. 127/NQ-CP on the application of electronic visas for citizens of countries and territories... Accordingly, the Government granted issueing electronic visas to citizens of all countries and territories." [sic 'issueing'] And: "The duration of Viet Nam's electronic visa is lengthened, from maximum of 30 days with single entry, to maximum of 90 days with single or multiple entry (ies)." PASSPORT-SET BASIS FOR "*": STRONG — the page affirmatively states the route is open to citizens of all countries and territories, so US, EU/EEA/CH, IN and CN passports are all covered by the source's own words, not by my inference. PERMITS: entry only; no work or study right attached. FAILS: 90 < 180 and no qualifying activity. This is the route relocation blogs most often mis-sell as nomad-compatible; it is not. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, https://vietnamembassy-usa.org/news/2023/10/new-policy-electronic-visa-and-visa-exemption fetched 2026-07-18. Both quotes verbatim, including the 'issueing' typo. The affirmative nationality-neutrality statement ('citizens of all countries and territories') is on the page as claimed — the STRONG basis for '*' holds. No work or study right stated; permits [VISIT] and maxStayDays 90 match. | seed | seed | 312b4f5c-0fc0-4ea5-848b-3072195d4bea | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 154 | 3b968669-61f2-4770-86c2-82b9338d2637 | vn | visa-access | {"route":"DH visa (study/internship) — Art. 8(12), Law on Entry/Exit (consolidated text 30/VBHN-VPQH)","permits":["RESIDE","STUDY"],"maxStayDays":365,"passportSet":"*"} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023) — Law 47/2014/QH13 as amended by Laws 51/2019/QH14 and 23/2023/QH15; rendering hosted by hethongphapluat.com | 2023-08-15 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the consolidated Art. 8/9/10: "12. DH - Cấp cho người vào thực tập, học tập." (issued to people entering for internship or study); "Thị thực ký hiệu NG1, NG2, NG3, NG4, LV1, LV2, ĐT4, DN1, DN2, NN1, NN2, NN3, DH, PV1, PV2, TT có thời hạn không quá 01 năm." (DH visa duration not exceeding 01 year → 365 days); condition: "Người nước ngoài vào học tập phải có văn bản tiếp nhận của nhà trường hoặc cơ sở giáo dục của Việt Nam." (study requires written acceptance of a Vietnamese school). WHAT IT PERMITS: presence (RESIDE, in the sense of lawful temporary stay) and STUDY at a Vietnamese institution. WHAT IT DOES NOT PERMIT: work — Art. 44 conditions employment on a work permit ("được lao động nếu có giấy phép lao động, trừ trường hợp không thuộc diện cấp giấy phép lao động"); the law is silent on remote work for a foreign employer, and silence is not permission. NEAREST MISS for the scenario: passes duration (365 ≥ 180) and two of three activities; fails WORK_REMOTE. PASSPORT-SET BASIS FOR "*": weaker basis — Art. 8 states the route with no nationality condition, and I did NOT find an affirmative nationality-neutrality statement for sponsored visa symbols; recording that distinction explicitly per PASSPORT.md §1. TIER DISCLOSURE: the instrument is the National Assembly Office's official consolidation; the URL is a commercial legal database rendering it, because every official channel for this text refused or defeated the fetcher (vbpl.vn ERR_CONNECTION_CLOSED; thuvienphapluat.vn and luatvietnam.vn Cloudflare 403; congbao.chinhphu.vn and vanban.chinhphu.vn serve PDF/viewer only). Tiered COMMERCIAL_LISTING for the URL actually read, not for the instrument. The official PDF of the newer consolidation 62/VBHN-VPQH (2026-03-23) is at https://vanban.chinhphu.vn/?docid=217302&pageid=27160 and should be read by a PDF-capable pass. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, hethongphapluat.com consolidated 30/VBHN-VPQH fetched 2026-07-18 (81,761 chars). All three claimed quotes plus the Art. 44 work-gating clause found verbatim. DH ≤ 01 year → 365 matches, and Art. 31 ('Thời hạn tạm trú cấp bằng thời hạn thị thực') makes it a single-route stay. RESIDE is additionally supported by DH's inclusion in the temporary-residence-card list (Art. 36: LV1, LV2, LS, ĐT1, ĐT2, ĐT3, NN1, NN2, DH, PV1, LĐ1, LĐ2, TT). No work right — correctly excluded. Tier honestly disclosed as COMMERCIAL_LISTING for the rendering. | seed | seed | 9a9edab9-22d1-4147-afbb-b9c25bd64fcf | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 155 | 23689d31-ec8b-44f3-ad8a-81c625e61cf4 | vn | visa-access | {"route":"LĐ1 visa (work, work-permit-exempt) — Art. 8(16), Law on Entry/Exit as amended by Law 51/2019/QH14","permits":["RESIDE","WORK_LOCAL"],"maxStayDays":730,"passportSet":"*"} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2020-07-01 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim: "LĐ1 - Cấp cho người nước ngoài làm việc tại Việt Nam có xác nhận không thuộc diện cấp giấy phép lao động, trừ trường hợp điều ước quốc tế mà Việt Nam là thành viên có quy định khác." (issued to foreigners working in Vietnam certified as work-permit-exempt); "Thị thực ký hiệu LĐ1, LĐ2 có thời hạn không quá 02 năm." (≤ 02 years → 730 days). Temporary residence card: "Thẻ tạm trú có ký hiệu LĐ1, LĐ2 và PV1 có thời hạn không quá 02 năm." WHAT IT PERMITS: residence and EMPLOYMENT IN VIETNAM for a Vietnamese-side employer/host (WORK_LOCAL). WHAT IT DOES NOT PERMIT: STUDY (separate condition — school acceptance, Art. 10) and WORK_REMOTE for a foreign employer, which the route is not shaped for: the symbol is defined by working 'tại Việt Nam' under Vietnamese labor law with a sponsoring entity. FAILS the scenario on activities despite passing duration (730 ≥ 180). PASSPORT-SET BASIS FOR "*": weaker basis — no nationality condition stated in Art. 8; no affirmative neutrality statement found. TIER DISCLOSURE: same as the DH claim — official consolidation, commercial rendering; official channels blocked to this fetcher. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, same consolidated text, fetched 2026-07-18. All three quotes verbatim. ≤ 02 years → 730 matches. WORK_LOCAL not WORK_REMOTE is right: the symbol is defined by working 'tại Việt Nam' with a Vietnamese-side exemption certification; nothing on the page grants remote work for a foreign employer. | seed | seed | 9ec8acc1-c385-463c-9384-675d746feec9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 156 | 5ac1e4aa-6a73-4a97-8a88-d1c6da2df157 | vn | visa-access | {"route":"LĐ2 visa (work, work-permit required) — Art. 8(16a), Law on Entry/Exit as amended by Law 51/2019/QH14","permits":["RESIDE","WORK_LOCAL"],"maxStayDays":730,"passportSet":"*"} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2020-07-01 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim: "LĐ2 - Cấp cho người nước ngoài làm việc tại Việt Nam thuộc diện phải có giấy phép lao động." (issued to foreigners working in Vietnam who require a work permit); "Thị thực ký hiệu LĐ1, LĐ2 có thời hạn không quá 02 năm." (≤ 02 years → 730 days). Rights clause, Art. 44: "được lao động nếu có giấy phép lao động, trừ trường hợp không thuộc diện cấp giấy phép lao động; được học tập nếu có văn bản tiếp nhận của nhà trường hoặc cơ sở giáo dục;" — work and study are each gated on their own separate authorization, which is why no symbol is fungible across purposes. WHAT IT PERMITS: residence plus employment in Vietnam under a work permit tied to a Vietnamese employer (WORK_LOCAL). WHAT IT DOES NOT PERMIT: STUDY as the entry purpose, and WORK_REMOTE for a foreign employer — a Vietnamese work permit presupposes a Vietnamese-side employer/sponsor. FAILS the scenario on activities despite passing duration. PASSPORT-SET BASIS FOR "*": weaker basis — no nationality condition stated in Art. 8; no affirmative neutrality statement found. TIER DISCLOSURE: as above. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, same consolidated text, fetched 2026-07-18. Quotes verbatim (16a carries footnote marker '[15]' in the rendering). 730 days and [RESIDE, WORK_LOCAL] match the source; the Art. 44 clause confirms work and study are separately gated. | seed | seed | 13f901e8-4c04-432e-84aa-c98e9a8c09cb | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 157 | 7de1e98e-d746-4169-b52b-98662356665b | vn | visa-access | {"route":"ĐT1 visa (investor, ≥ VND 100bn) — Art. 8(7a), Law on Entry/Exit as amended by Law 51/2019/QH14","permits":["RESIDE","INVEST"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://tuyentruyenphapluat.tphcm.gov.vn/index.php/outlinelegaldoc/145/n%E1%BB%99i-dung-c%C6%A1-b%E1%BA%A3n-c%E1%BB%A7a-lu%E1%BA%ADt-s%E1%BB%ADa-%C4%91%E1%BB%95i-b%E1%BB%95-sung-m%E1%BB%99t-s%E1%BB%91-%C4%91i%E1%BB%81u-c%E1%BB%A7a-lu%E1%BA%ADt-nh%E1%BA%ADp-c%E1%BA%A3nh-xu%E1%BA%A5/ | Cổng thông tin Tuyên truyền phổ biến pháp luật Thành phố Hồ Chí Minh (Ho Chi Minh City legal dissemination portal) | 2020-07-01 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the HCMC government portal's article on Law 51/2019: "7a. ĐT1 - Cấp cho nhà đầu tư nước ngoài tại Việt Nam và người đại diện cho tổ chức nước ngoài đầu tư tại Việt Nam có vốn góp giá trị từ 100 tỷ đồng trở lên hoặc đầu tư vào ngành, nghề ưu đãi đầu tư, địa bàn ưu đãi đầu tư do Chính phủ quyết định." and "nhà đầu tư (ĐT1) có vốn góp giá trị từ 100 tỷ đồng trở lên thị thực có thời hạn không quá 05 năm và được cấp thẻ tạm trú đến 10 năm" (visa ≤ 05 years → 1825 days; temporary residence card up to 10 years). The same page states the full series: ĐT2 (VND 50–<100bn, visa ≤ 5yr), ĐT3 (VND 3–<50bn, visa ≤ 3yr, TRC ≤ 3yr), ĐT4 (< VND 3bn, visa ≤ 12 months) — quoted in researcher-notes; not filed as separate claims because none changes the verdict and each shares this one source. WHAT IT PERMITS: residence as an investor (RESIDE + INVEST, capital ≥ VND 100bn ≈ USD 4m, purpose-bound to an investment in Vietnam under the Law on Investment — Art. 10 requires 'giấy tờ chứng minh việc đầu tư tại Việt Nam theo quy định của Luật đầu tư'). WHAT IT DOES NOT PERMIT: STUDY, and WORK_REMOTE for a foreign employer — both unstated, and silence is not permission. FAILS the scenario on activities despite the longest duration in the ordinary system. PASSPORT-SET BASIS FOR "*": weaker basis — no nationality condition stated; no affirmative neutrality statement found. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, tuyentruyenphapluat.tphcm.gov.vn fetched 2026-07-18. Both quotes verbatim; the ĐT2/ĐT3/ĐT4 series is on the page as described. Visa ≤ 05 years → 1825 matches (independently corroborated by the consolidated law's 'Thị thực ký hiệu LS, ĐT1, ĐT2 có thời hạn không quá 05 năm'). No study or remote-work right stated; [RESIDE, INVEST] is the correct reading. | seed | seed | d04fa8b0-2e85-4222-9d8b-541b7f6a2a50 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 158 | c6bb1ab6-bee8-4034-801b-572fa830934c | vn | visa-access | {"route":"VR visa (visiting relatives / other purposes) — Art. 8(19), Law on Entry/Exit (consolidated text 30/VBHN-VPQH)","permits":["VISIT"],"maxStayDays":180,"passportSet":"*"} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2023-08-15 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim: "19. VR - Cấp cho người vào thăm người thân hoặc với mục đích khác." (issued to people entering to visit relatives or for other purposes); "Thị thực ký hiệu VR có thời hạn không quá 180 ngày." (≤ 180 days). WHAT IT PERMITS: entry and stay only (VISIT). It is the one ordinary symbol that exactly meets the 180-day bar — and it is the row that would have tripped the pre-D-025 tourist-duration bug: duration passes, purpose fails. WHAT IT DOES NOT PERMIT: RESIDE as a status, STUDY, WORK_REMOTE — none stated; Art. 44 gates work and study on separate authorizations. FAILS the scenario on all three intended activities. PASSPORT-SET BASIS FOR "*": weaker basis — no nationality condition stated in Art. 8; no affirmative neutrality statement found. TIER DISCLOSURE: as on the DH claim. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, same consolidated text, fetched 2026-07-18. Both quotes verbatim. 180 days matches and is a single-route stay (Art. 31: temporary stay certificate equals visa duration; the 30-day carve-out applies only to DL). [VISIT] only — no work/study/residence right stated. The duration-passes/purpose-fails reading is exactly what the source supports. | seed | seed | 40b5f168-2437-4789-9145-de265bd8b047 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 159 | 625d387f-fdc0-45b2-b229-973527a2d73f | vn | visa-access | {"route":"DL visa (tourist) — Art. 8(17), Law on Entry/Exit (consolidated text 30/VBHN-VPQH)","permits":["VISIT"],"maxStayDays":30,"passportSet":"*"} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2023-08-15 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim: "17. DL - Cấp cho người vào du lịch." (issued to tourists); "Thị thực ký hiệu HN, DL, EV có thời hạn không quá 90 ngày." (≤ 90 days). CAUTION ON STAY-PER-ENTRY: the amended Art. 31 (per the HCMC portal's summary of Law 51/2019, quoted in researcher-notes) grants a DL visa holding more than 30 days only a 30-day temporary stay certificate per entry, extension discretionary — so the practical single-entry stay is 30 days even though the visa runs to 90. I record maxStayDays 90 as the visa's own stated ceiling and flag the 30-day stay rule rather than silently choosing either number. WHAT IT PERMITS: tourism only (VISIT). FAILS the scenario on both duration and every intended activity. PASSPORT-SET BASIS FOR "*": weaker basis — no nationality condition stated; no affirmative neutrality statement found. TIER DISCLOSURE: as on the DH claim. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CORRECTED, and the value above IS the corrected one (maxStayDays 30, not the 90 the researcher recorded). VERIFIER: OK 200, same consolidated text, fetched 2026-07-18. The two claimed quotes ('17. DL - Cấp cho người vào du lịch.' and 'Thị thực ký hiệu HN, DL, EV có thời hạn không quá 90 ngày.') are verbatim, and the researcher flagged the 30-day rule honestly. But the cited source ITSELF states the Art. 31(1)(a) rule quoted here: a DL visa over 30 days yields a 30-day temporary stay per entry, with extension expressly discretionary ('được xem xét'). The claim set's own rule — applied by the researcher to the SVEC claim (card 5 years, maxStayDays recorded as the 90-day per-entry stay because extension is discretionary) — requires the same treatment here: single-route stay is 30, not the 90-day visa validity. 90 counts a discretionary extension. Correction changes no scenario verdict (30 and 90 both fail the 180-day bar), but maxStayDays should be 30 for consistency with how the researcher scored SVEC and with the per-entry-stay definition. CORRECTING QUOTE, verbatim: “Thời hạn tạm trú cấp bằng thời hạn thị thực; trường hợp thị thực có ký hiệu DL thời hạn trên 30 ngày thì cấp tạm trú 30 ngày và được xem xét gia hạn tạm trú theo quy định tại Điều 35 của Luật này” | seed | seed | 4c51901a-8c0c-480a-b722-9d5361d274df | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 160 | 1bf2b9bc-6822-411c-8fd5-87ea7e05115b | vn | visa-access | {"route":"UĐ1 visa (high-quality digital-technology-industry personnel and preferential categories) — Art. 8(22), added by Law 118/2025/QH15","permits":["RESIDE"],"maxStayDays":1825,"passportSet":"*"} | NULL | https://hvannd.edu.vn/mot-so-diem-moi-cua-luat-so-1182025qh15-sua-doi-bo-sung-mot-so-dieu-cua-10-luat-co-lien-quan-den-an-ninh-trat-tu | Học viện An ninh nhân dân (People's Security Academy, Ministry of Public Security) | 2026-07-01 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTE, verbatim from the MPS academy's article (re-fetched 2026-07-18; the article is dated 07/05/2026 and states the law was promulgated 10/12/2025): "Tại khoản 2 Điều 2, Luật bổ sung khoản 22 và khoản 23 vào Điều 8 của Luật Nhập cảnh, xuất cảnh, quá cảnh, cư trú của người nước ngoài tại Việt Nam, gồm: (1) Thị thực ký hiệu UĐ1: cấp cho người nước ngoài là nhân lực công nghiệp công nghệ số chất lượng cao và các trường hợp được ưu đãi theo quy định; (2) Thị thực ký hiệu UĐ2: cấp cho vợ, chồng, con dưới 18 tuổi của người nước ngoài được cấp thị thực UĐ1; (3) quy định thị thực ký hiệu LS, ĐT1, ĐT2, UĐ1, UĐ2 có thời hạn không quá 05 năm (khoản 3 Điều 2 sửa đổi, bổ sung khoản 6 Điều 9); quy định thẻ tạm trú ký hiệu ĐT1, UĐ1, UĐ2 có thời hạn không quá 10 năm (Khoản 10 Điều 2 sửa đổi, bổ sung khoản 2 Điều 38)." So: visa ≤ 5 years (1825 days), TRC ≤ 10 years. WHAT IT PERMITS: residence for a person who IS high-quality digital-technology-industry personnel under the Law on Digital Technology Industry, or in a legally-designated preferential category. Employment still runs through the work-permit regime (Art. 44, unamended on this point) and the category is Vietnam-industry-bound; nothing here permits STUDY or WORK_REMOTE for a foreign employer. This is the strongest new route in the system and it still fails the scenario on activities. EFFECTIVE DATE BASIS: the page states the child-seat provision's 2026-07-01 effective date via Art. 11 of Law 118/2025; multiple official republications state the law's general effective date as 01/07/2026 — I did not find a source stating a different date for the visa articles. SOURCING GAP, disclosed: the National Assembly's own consolidated text (VBHN 62/VBHN-VPQH, 2026-03-23) exists at https://vanban.chinhphu.vn/?docid=217302&pageid=27160 but only as a PDF this fetcher cannot render; the official Law 118/2025 PDF at datafiles.chinhphu.vn also failed as a download. A PDF-capable pass should read either to upgrade this claim to OFFICIAL_PRIMARY. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, hvannd.edu.vn (article dated 07/05/2026, law promulgated 10/12/2025) fetched 2026-07-18. Quote verbatim in full. Visa ≤ 05 years → 1825 matches; [RESIDE] only is the correct conservative reading — the page states no work or study right, and the article confirms Art. 44 is untouched by Law 118/2025's entry/exit amendments as summarized. The 2026-07-01 effective date is stated on the page only for the child-seat provision via Art. 11 of Law 118/2025, exactly as the claim discloses; the visa-article date rests on the disclosed secondary basis and the OFFICIAL_PRIMARY upgrade path (VBHN 62 PDF) is correctly flagged. | seed | seed | c6ca37b2-860f-46b0-b5e3-5e6e9952c0ee | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 161 | d2a1340e-b933-4f4a-b240-b46755e3892f | vn | visa-access | {"route":"Special visa-exemption card (thẻ miễn thị thực đặc biệt, SVEC) — Decree 221/2025/NĐ-CP","permits":["VISIT"],"maxStayDays":90,"passportSet":"*"} | NULL | https://xaydungchinhsach.chinhphu.vn/nghi-dinh-so-221-2025-nd-cp-mien-thi-thuc-co-thoi-han-cho-nguoi-nuoc-ngoai-thuoc-dien-doi-tuong-dac-biet-119250809110419124.htm | Báo Điện tử Chính phủ / Cổng Thông tin điện tử Chính phủ (Government of Vietnam policy portal) | 2025-08-15 | 2026-07-18T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | SOURCE QUOTES, verbatim from the Government portal's article on Decree 221/2025/NĐ-CP (dated 08/08/2025): "Người nước ngoài sử dụng thẻ miễn thị thực đặc biệt được cấp chứng nhận tạm trú 90 ngày cho mỗi lần nhập cảnh, trường hợp thời hạn thẻ miễn thị thực đặc biệt ngắn hơn 90 ngày thì cấp chứng nhận tạm trú bằng thời hạn thẻ miễn thị thực đặc biệt; nếu có nhu cầu tiếp tục ở lại Việt Nam thì được xem xét gia hạn tạm trú hoặc cấp thị thực hoặc cấp thẻ tạm trú theo quy định..." and "Thời hạn miễn thị thực không quá 05 năm và ngắn hơn thời hạn còn lại của hộ chiếu ít nhất 30 ngày." So: card up to 5 years, multiple entry, but 90 DAYS of temporary residence per entry, with extension expressly discretionary (được xem xét) — per the gate's rule the discretionary extension must not be counted, so maxStayDays is 90 and fails the 180-day bar. ELIGIBILITY, quoted: investors/leaders must belong to a company in "danh sách 100 doanh nghiệp có giá trị vốn hóa lớn nhất thế giới" (world's top-100 by market cap); scientists must hold "giải thưởng quốc tế uy tín" (a recognized prestigious international award); every substantive category requires "văn bản đề nghị của bộ, cơ quan ngang bộ, cơ quan thuộc Chính phủ hoặc cơ quan Đảng, Quốc hội" (a written request from a ministry or Party/NA agency). WHAT IT PERMITS: entry exemption only (VISIT) — it confers no work or study right. This kills the 'Vietnam golden visa' story as a long-stay route. PASSPORT-SET BASIS FOR "*": weaker basis — the decree's categories are institutional/elite with no nationality condition stated; no affirmative neutrality statement found. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, xaydungchinhsach.chinhphu.vn (article dated 09/08/2025 on Decree of 8/8/2025) fetched 2026-07-18. All claimed quotes verbatim, including top-100-market-cap and international-award criteria. maxStayDays 90 correctly records the per-entry stay, not the 5-year card, with the extension discretionary ('được xem xét') — the right rule, and the benchmark against which claim 7 was corrected. One minor note-level overstatement that does not touch the value: the ministry/Party/NA written-request condition applies to categories b, c, d, đ; categories e and g instead require 'văn bản đề nghị của cơ quan, tổ chức có liên quan' (a request from the relevant agency/organization). Every category still requires an institutional sponsor, so the substance — no self-service route — stands. | seed | seed | eec7c11f-4c66-41f3-9afc-a6dee8855ff5 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 162 | caa30b43-2d5f-4b6e-8ae3-1e081be9cef3 | vn | visa-access | {"passportSet":"US","routesEnumerated":true} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2026-07-01 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | ENUMERATION CLAIM. Vietnam's visa system is a closed, purpose-bound enumeration: Art. 8 ('Điều 8. Ký hiệu thị thực') lists every visa symbol that exists and Art. 9 caps each symbol's duration. I read the complete consolidated Art. 8 at this URL — items 1 through 21 (NG1, NG2, NG3, NG4, LV1, LV2, ĐT1, ĐT2, ĐT3, ĐT4 [7a–7d], DN1, DN2 [8], NN1, NN2, NN3, DH, HN, PV1, PV2, LĐ1, LĐ2 [16, 16a], DL, TT, VR, SQ, EV [21]) — plus the only post-2023 additions, items 22–23 (UĐ1, UĐ2), added by Law 118/2025/QH15 per the MPS academy source cited on the UĐ1 claim ('Luật bổ sung khoản 22 và khoản 23 vào Điều 8'). Art. 8 states every route without any nationality condition, so the same closed list is the complete route list for a US passport. NATIONALITY-SPECIFIC OVERLAYS CHECKED for US: the unilateral 45-day visa exemption (Resolution 128/NQ-CP, quoted in researcher-notes) names 13 countries and the United States is NOT among them; no US-specific route or exemption was found, and the e-visa (open to all nationalities) is already claimed. CONCLUSION THE GATE MAY DRAW: no symbol in the complete list permits RESIDE+STUDY+WORK_REMOTE together, and the near-misses are stated per route above. CAVEATS, disclosed: (1) the single document read here consolidates through Law 23/2023; the UĐ1/UĐ2 additions were read in a separate official-secondary source, so this enumeration spans two cited documents rather than one; (2) the newer consolidation VBHN 62/VBHN-VPQH (2026-03-23, post-118/2025) exists at vanban.chinhphu.vn docid=217302 as an unread PDF — reading it would collapse this to one document; (3) routes created outside the Law were checked: Decree 221/2025 (claimed above, fails), coastal economic zone entry exemption (Art. 12(3a), 30 days, Phú Quốc-type — entry only), and bilateral treaty exemptions for US ordinary passports (none found). ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, all three underlying pages re-fetched 2026-07-18. This is a real D-036 enumeration, not an 'I could not find one': the cited page carries the complete closed Art. 8 list, verified to run 1 through 21 and terminate at Điều 9 (start '1. NG1', middle items 7a-7d/12/13/14/16/16a/17/18/19 all present, end '20. SQ', '21.[17] EV'); the only post-2023 additions (22-23, UĐ1/UĐ2) are verified verbatim in the claim set's cited MPS academy source; and the 13-state 45-day exemption list was independently re-read on the embassy page — the United States is not on it. Art. 8 states no nationality condition on any symbol. Caveat stands as disclosed: the enumeration spans two cited documents, and the unread VBHN 62 PDF is the single-document upgrade path. | seed | seed | c9069236-e24b-4678-a4b6-ea56d3d1d441 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 163 | 4594e821-69c9-4a57-86ea-c8bcc267929d | vn | visa-access | {"passportSet":"EU/EEA/CH","routesEnumerated":true} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2026-07-01 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | ENUMERATION CLAIM, with an identifier caveat stated up front: Vietnamese law does NOT use the bloc 'EU/EEA/CH' — no Vietnamese instrument states one rule for that bloc. This key is the SCENARIO'S identifier (scoring.config.json passportSets), used solely so the enumeration is addressable to the cohort; the substantive basis is that Art. 8's closed route list (items 1–21 read in full at this URL, items 22–23 UĐ1/UĐ2 added by Law 118/2025 per the MPS academy source) is stated without any nationality condition, so it is the complete route list for every EU/EEA/CH passport holder identically. NATIONALITY-SPECIFIC OVERLAYS CHECKED: the unilateral 45-day exemption (Resolution 128/NQ-CP as quoted verbatim by the Vietnam Embassy in the US, cited in researcher-notes) names SEVEN EU member states individually — Germany, France, Italy, Spain, Denmark, Sweden, Finland (the other six named states are non-EU: UK, Russia, Japan, Korea, Norway [EEA], Belarus). It does not cover the EU bloc uniformly, it is 45 days (< 180), and it is an entry exemption 'regardless of passport type, purpose of entry' that confers no work or study right — so it cannot create a qualifying route for any member of this passport set, covered or not. It is deliberately NOT filed as a route claim keyed 'EU/EEA/CH' because the source does not name that bloc, and splitting it into 13 per-country rows would add no route relevant to the scenario's passport sets; the full quote and list are in researcher-notes. No other EU/EEA/CH-specific instrument was found (2025 tourism-tied exemptions for Poland/Czechia/Switzerland are noted in researcher-notes as leads read only in secondary reporting). Same two-document and unread-VBHN-62 caveats as the US enumeration. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, underlying pages re-fetched 2026-07-18. Same complete Art. 8 basis as claim 10, verified. The 13-state list checks out exactly as the claim counts it: 7 EU members named individually (Germany, France, Italy, Spain, Denmark, Sweden, Finland) and 6 non-EU (UK, Russia, Japan, Korea, Norway, Belarus); 45 < 180 and entry-only, so it cannot create a qualifying route for any cohort member, covered or not. IDENTIFIER CAVEAT (flagged, not a refutation): 'EU/EEA/CH' is not a bloc any cited Vietnamese source names — PASSPORT.md §1 permits blocs only as quotations. The claim discloses this openly and grounds the enumeration in Art. 8's nationality-unconditional list, which does apply to every member of the set identically; but the parent should note this key is a scenario identifier riding on a '*'-shaped basis, not a source-quoted bloc. | seed | seed | 89c3ca91-6992-4393-8d12-d8d336aa6406 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 164 | 29625ac4-cb14-42f6-9e54-4dbce9118f31 | vn | visa-access | {"passportSet":"IN","routesEnumerated":true} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2026-07-01 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | ENUMERATION CLAIM. Same basis as the US enumeration: Art. 8's closed list (items 1–21 read in full at this URL; items 22–23 UĐ1/UĐ2 added by Law 118/2025 per the MPS academy source) is stated without any nationality condition, so it is the complete route list for an Indian passport. NATIONALITY-SPECIFIC OVERLAYS CHECKED for IN: India is NOT among the 13 states named in the unilateral 45-day exemption (Resolution 128/NQ-CP, quoted in researcher-notes); no India-specific route, exemption, or restriction was found in any source read this pass. Indian passport holders reach Vietnam through the universal e-visa (claimed above, 90 days, entry only) or the sponsored symbols of Art. 8. No symbol permits RESIDE+STUDY+WORK_REMOTE together. Same two-document and unread-VBHN-62 caveats as the US enumeration. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, underlying pages re-fetched 2026-07-18. Same verified complete-list basis as claim 10. India is not on the 45-day list; the e-visa's all-nationalities coverage (Resolution 127, claim 1's page) was independently confirmed, so the IN route set is exactly the universal set — no overlay found on any page read, and none is asserted. | seed | seed | df7ff565-71aa-45ea-aef2-b06b140510b4 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 165 | 7f2fb308-3495-44dc-915b-39be3cb39a29 | vn | visa-access | {"passportSet":"CN","routesEnumerated":true} | NULL | https://hethongphapluat.com/van-ban-hop-nhat-30-vbhn-vpqh-nam-2023-hop-nhat-luat-nhap-canh-xuat-canh-qua-canh-cu-tru-cua-nguoi-nuoc-ngoai-tai-viet-nam-do-van-phong-chinh-phu-ban-hanh.html | Văn phòng Quốc hội, Văn bản hợp nhất 30/VBHN-VPQH (2023); rendering hosted by hethongphapluat.com | 2026-07-01 | 2026-07-18T00:00:00.000Z | COMMERCIAL_LISTING | CORROBORATED | MODEL_ASSISTED_RESEARCH | NULL | ENUMERATION CLAIM. Same basis as the US enumeration: Art. 8's closed list (items 1–21 read in full at this URL; items 22–23 UĐ1/UĐ2 added by Law 118/2025 per the MPS academy source) is stated without any nationality condition, so it is the complete route list for a Chinese passport. NATIONALITY-SPECIFIC OVERLAYS CHECKED for CN: China is NOT among the 13 states named in the unilateral 45-day exemption (Resolution 128/NQ-CP); no CN-specific exemption for ORDINARY passports was found, and no CN-specific restriction or exclusion was found either — the e-visa is affirmatively open to all countries and territories (Resolution 127/NQ-CP, claimed above), and Resolution 127's border-gate list includes the land gates on the China border (Móng Cái, Hữu Nghị, Lào Cai), consistent with no CN carve-out. I did NOT verify the Vietnam–China bilateral agreement on mutual exemption for diplomatic/official passports because it cannot cover an ordinary-passport cohort; recorded as a lead only. No symbol permits RESIDE+STUDY+WORK_REMOTE together. Same two-document and unread-VBHN-62 caveats as the US enumeration. ADVERSARIALLY VERIFIED (D-032), 2026-07-18 — verdict CONFIRMED by an independent agent that had not seen the research and re-fetched the source with the same instrument. VERIFIER: OK 200, underlying pages re-fetched 2026-07-18. Same verified complete-list basis as claim 10. No CN-specific exemption or restriction for ordinary passports appears on any cited page; the diplomatic/official-passport bilateral is correctly left as a lead, not a claim. | seed | seed | 66b2876d-5eaf-4bc6-95aa-cf082257e0a4 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 166 | 7446b598-3326-4062-937c-5357a038d55c | vn | operating-licence | true | NULL | https://en.baochinhphu.vn/regulations-on-foreign-cooperation-investment-in-education-11133384.htm | Vietnam Government Portal (baochinhphu.vn) | 2018-08-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | DOCUMENT_EXTRACTION | NULL | TRUE — a foreign entity MAY lawfully run an educational institution in Vietnam. This is a real, enumerated licensing regime, not an inference: Decree 86/2018/NĐ-CP governs foreign cooperation and investment in education and expressly permits foreign-invested educational institutions, defined as 'institutions financed by foreign invested business entities which operate in accordance with the laws of Vietnam and are allowed to use their own seals and accounts'. The quoted list of permitted FORMS is the operative fact — a Network School would have to fit one of them, most plausibly 'Short-term training and retraining institutions'. IMPORTANT QUALIFIERS, none of which negate the gate but all of which bear on feasibility: (a) it is a permission regime, not a right — establishment requires Prime Ministerial approval (Art. 40), with the Education Minister or provincial People's Committee competent in specific cases, and a separate operational licence (Arts. 45-47); (b) minimum investment capital for a higher-education institution runs VND 300bn-1,000bn; (c) Vietnamese students on a foreign programme must stay below 50% of that programme's enrolment — a real constraint on a mixed-nationality cohort; (d) operation is capped at 50 years and cannot exceed the property lease term; (e) vocational education is carved OUT of this decree and sits under a different regime. Corroborated across the Government portal and the decree text. CAVEAT: Decree 86/2018 has been amended (Decree 124/2024/NĐ-CP is reported to amend it); I did not verify the current consolidated text, so treat thresholds as indicative. SOURCE QUOTE, verbatim from the page on the retrieval date: “Short-term training and retraining institutions; Nursery education institutions; High-school education institutions (elementary schools, middle schools, high schools, and mixed level schools); Higher education institutions; Campuses of foreign-invested higher education institutions in Viet Nam. [...] the minimum investment capital for projects of establishment of higher education institutions from VND 300 billion to VND 1000 billion [...] Vietnamese students attending foreign educational programs must be lower than 50% of the total number of students attending foreign educational programs” | seed | seed | f0d87abd-aef8-48cc-98d1-d6cb3ed98d2a | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 167 | eb0cbb56-b4e4-43ff-b6ae-da71acee9e8e | vn | political-stability | 65.9913074 | WGI percentile 0-100 | https://api.worldbank.org/v2/country/VNM/indicator/GOV_WGI_PV.SC?format=json&source=3&date=2020:2025 | World Bank, Worldwide Governance Indicators | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | ANALYST_VERIFIED | OFFICIAL_API | NULL | Vietnam, Political Stability and Absence of Violence/Terrorism, 2024 (latest available). Higher is better. Value returned live by the World Bank API. NOTE FOR THE PIPELINE: the legacy WGI indicator codes (PV.PER.RNK etc.) are now DEAD on the WB API — they return 'The indicator was not found. It may have been deleted or archived.' WGI has been restructured under source=3 with codes GOV_WGI_PV.SC / GOV_WGI_RL.SC, where '.SC' is the 0-100 governance score (the percentile-rank equivalent this criterion wants). Any part of this product still querying the old codes is silently getting nothing. Trend is mildly negative: 68.12 (2020) → 68.19 (2021) → 67.99 (2022) → 67.44 (2023) → 65.99 (2024). SOURCE QUOTE, verbatim from the page on the retrieval date: “2024 65.9913074 | Political Stability - Governance score (0-100)” | seed | seed | 09d2c906-ddbb-453f-9564-29c0856e0e59 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 168 | 3fdb67a6-29a9-4c41-baee-5c2329cdf3a1 | vn | rule-of-law | 51.5616803 | WGI percentile 0-100 | https://api.worldbank.org/v2/country/VNM/indicator/GOV_WGI_RL.SC?format=json&source=3&date=2020:2025 | World Bank, Worldwide Governance Indicators | 2024-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | ANALYST_VERIFIED | OFFICIAL_API | NULL | Vietnam, Rule of Law, 2024 (latest available). Higher is better. Returned live by the World Bank API under the restructured source=3 code (see political-stability note — the legacy RL.PER.RNK code is dead). Essentially flat for five years: 51.77 (2020) → 51.62 (2021) → 51.17 (2022) → 51.11 (2023) → 51.56 (2024). Roughly the global median, and a material input to the visa/licence findings above: Vietnam's regimes are written down and enumerable, but discretionary approval (PM sign-off for schools, institutional nomination for UĐ1, discretionary stay extensions) recurs throughout. SOURCE QUOTE, verbatim from the page on the retrieval date: “2024 51.5616803 | Rule of Law - Governance score (0-100)” | seed | seed | 83dbb475-fcd7-4ece-9b88-fc58d57292e6 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 169 | d2504897-8b0a-411f-ba48-7bf990cf616e | vn | corruption | 41 | TI CPI 0-100 | https://www.transparency.org/en/countries/vietnam | Transparency International | 2025-12-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | ANALYST_VERIFIED | DOCUMENT_EXTRACTION | NULL | Vietnam CPI = 41/100, ranked 81 of 182, latest edition as displayed on TI's Vietnam country page (CPI 2025, published early 2026). Higher is better. Up 1 point year-on-year. TI CPI is explicitly permitted for this product. The page renders the figure as 'this year' rather than naming the edition, so effectiveAt is set to the 2025 reference year on the basis that TI publishes the CPI for year N in early N+1 and today is 2026-07-17; if the pipeline needs the edition named explicitly, re-verify against TI's CPI 2025 release page. SOURCE QUOTE, verbatim from the page on the retrieval date: “Vietnam has a score of 41 this year, with a change of 1 since last year, meaning it ranks 81 out of 182 countries.” | seed | seed | 9a6e55fb-b9f3-41d9-8b8b-b46e65bf285c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 170 | 30eab3ba-41fa-48f7-93aa-3793970ed0ad | vn | currency-inflation | 3.3 | CPI %/yr | https://www.imf.org/external/datamapper/api/v1/PCPIPCH/VNM | International Monetary Fund, World Economic Outlook | 2025-12-31 | 2026-07-17T00:00:00.000Z | MULTILATERAL | ANALYST_VERIFIED | OFFICIAL_API | NULL | Vietnam, WEO indicator PCPIPCH (Inflation, average consumer prices, annual % change), 2025 = 3.3. Lower is better; this is a strong result, comfortably inside the 0-50 band and low by regional standards. Series retrieved: 2022=3.2, 2023=3.3, 2024=3.6, 2025=3.3, 2026=4.9, 2027=4.6. I report the 2025 value as the latest reference year at or before retrievedAt; the 2026 and 2027 figures are WEO PROJECTIONS and are deliberately NOT used, but flag that the projection rises to 4.9 for 2026 — a ~48% step up. CAVEAT ON METHOD: the datamapper endpoint IGNORES the country path segment and returns the full all-country payload; VNM must be parsed out of values.PCPIPCH.VNM. Fetching that URL and reading the first country returned (SDN/Sudan, 100.2 for 2025) would produce a catastrophically wrong number with a real URL attached. Any automated collector hitting this endpoint must parse by key, never by position. SOURCE QUOTE, verbatim from the page on the retrieval date: “2022 3.2 / 2023 3.3 / 2024 3.6 / 2025 3.3 / 2026 4.9 / 2027 4.6” | seed | seed | 7a8d5586-18b6-4b8f-8deb-87d62836a4e8 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 171 | 7ba1d30c-ff94-48b7-9d0b-fc06aa3f53b5 | vn | english-proficiency | 500 | EF EPI score 300-700 | https://www.ef.com/wwen/epi/regions/asia/vietnam/ | EF Education First, English Proficiency Index | 2025-12-31 | 2026-07-17T00:00:00.000Z | ACADEMIC_OR_INDEX | UNVERIFIED | DOCUMENT_EXTRACTION | NULL | Vietnam EF EPI = 500. EF is the only source for this criterion, as the brief states, and its licensing for commercial use is UNRESOLVED — flagging that explicitly per the brief's instruction. Edition year is 500's weak point: Vietnam's own country page does NOT print an edition year (its only date is a generic '© Signum International AG 2026' footer), so I fetched the EPI index root separately, which returns '2025 Edition'; effectiveAt is set from that, not from the country page. Marked UNVERIFIED rather than ANALYST_VERIFIED because the score and the edition year come from two different pages and I could not see them asserted together. 500 sits at the boundary of EF's 'Moderate' band. Practical read for site selection: adequate for an English-medium operation but well below Singapore/Malaysia. SOURCE QUOTE, verbatim from the page on the retrieval date: “500” | seed | seed | 24ee5108-3d77-4e99-b8b8-03fdd42a3fa4 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 172 | b3eabbbc-4ffd-448e-8142-30d877029fd4 | vn | entry-friction | UNRESOLVED | days | https://hanoi-be.mfa.gov.tr/Content/assets/consulate/images/localCache/1/354983aa-6b1a-4df5-b559-e94969bfc05c.pdf | National Assembly of Vietnam, Law No. 47/2014/QH13 (copy hosted by the Turkish Embassy in Hanoi) | 2015-01-01 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | DEGRADED BY ADVERSARIAL VERIFICATION (D-032). The researcher claimed 12 and an independent agent re-fetched the source and rejected it. VERIFIER: URL resolves and both quoted sentences are verbatim in the extracted PDF: Art. 16(3) '3. Within 05 working days from the receipt of the application for the visa, the immigration authority shall consider granting it...' (line 253) and Art. 17(2) 'Within 03 working days from the receipt of the notification from the immigration authority or a competent authority of the Ministry of Foreign Affairs, the overseas visa-issuing authority of Vietnam shall issue the visa.' (line 276). But the quote does NOT establish the value 12. The two cited steps sum to 8 working days (5 + 3), not 12. Nothing in the quoted text yields 12 by any stated arithmetic; the document's only '12' near this topic is '12 working hours' for certain border-checkpoint cases (Art. 16(5)), which is not in the claimed quote and would not produce 12 days either. Either the value is a rubric score the page never states, or it is an arithmetic error — either way the quote does not support it. ORIGINAL RESEARCH NOTE, retained because the sourcing work is still worth having: DERIVED, NOT OBSERVED — read the conversion before using. CONVERSION: the gate's qualifying-shaped route is the sponsored visa (DH is the near-miss), so I priced that path, not the e-visa. Statutory ceilings, both quoted: 5 working days for the immigration authority to consider and respond to the inviting entity (Art. 16(3)) + 3 working days for the overseas mission to issue once notified (Art. 17(2)) = 8 working days. 8 working days ÷ 5 per week × 7 = 11.2 calendar days → rounded to 12. ASSUMPTIONS, all of which could move this: (a) these are statutory MAXIMA, not observed medians — actual practice may be faster; (b) the clock excludes the applicant-side work of obtaining the sponsoring entity's invitation and, for DH, the school's written acceptance required by Art. 10(4)(d), which is unbounded in the statute and realistically dominates the true elapsed time — so 12 is a FLOOR, and the honest expectation is materially worse; (c) COHORT-WEIGHTING IS A NO-OP: Art. 8 draws no nationality distinction, so the same figure applies at 30/25/25/20 and the weighted value equals the unweighted one. UNVERIFIED because no official source states an end-to-end elapsed time; the Immigration Department's own /instruction page, which would carry the e-visa processing time, returned a permission error (see blocked). QUOTE THE RESEARCHER OFFERED: “3. Within 05 working days from the receipt of the application for the visa, the immigration authority shall consider granting it, send a response to the inviting entity, and notify the overseas visa-issuing authority of Vietnam. [...] Within 03 working days from the receipt of the notification from the immigration authority or a competent authority of the Ministry of Foreign Affairs, the overseas visa-issuing authority of Vietnam shall issue the visa.” | seed | seed | a5719717-a17f-4cd5-81ae-93dd538b0497 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 173 | d76f414c-5206-40d0-a1a9-841813e8a656 | vn | government-receptivity | 2 | rubric 0-5 | https://internationaleducation.gov.au/International-network/vietnam/publications/Documents/Overview%20of%20Decree%2086.pdf | Australian Government Department of Education and Training | 2024-11-20 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | UNVERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 3: Named policy fit: a programme this project class demonstrably qualifies for, published eligibility criteria (SEZ regime, education-hub licence, digital-nomad statute). THE FACTS: · Vietnam operates a published licensing regime for foreign-invested education institutions, including short-course training institutions — the category a Network School node would enter under. Approval authority is named per category. https://internationaleducation.gov.au/International-network/vietnam/publications/Documents/Overview%20of%20Decree%2086.pdf “Foreign-invested education institutions under Decree 86 refer to institutions delivering short course training, kindergartens, schools, universities and branches of foreign-invested universities in Vietnam. The licencing process for foreign-invested education institutions involves three steps (outlined below), requiring approval from different authorities.” · Eligibility criteria are published and quantified — an explicit capital threshold per institution type, i.e. a rule-based regime rather than a discretionary welcome. https://internationaleducation.gov.au/International-network/vietnam/publications/Documents/Overview%20of%20Decree%2086.pdf “Minimum investment capital for (i) A foreign invested university: VND1000 billions (USD$45 million) which is tripled over previous requirement of VND300 billions (USD$15 millions), exclusive of land expenditure (ii) branch of a foreign invested university: VND250 billions (USD$11.5 million), exclusive of land expenditure” · The regime is current, not historical: Decree 86/2018/ND-CP was amended by Decree 124/2024/ND-CP, in force since November 2024. https://opportunities-insight.britishcouncil.org/short-articles/news/viet-nam-updates-regulations-foreign-education-cooperation “was passed by the Deputy Prime Minister in October 2024 and took effect in November 2024” REMAPPED BY ADVERSARIAL VERIFICATION (D-032/D-035). The scorer claimed 3 at "anchor 3: Named policy fit: a programme this project class demonstrably qualifies for, published eligibility criteria (SEZ regime, education-hub licence, digital-nomad statute).". An independent agent re-checked the facts against the pages AND the mapping against contracts/rubric.md, and the facts do not reach that anchor. VERIFIER: FACTS: Re-fetched the Decree 86 PDF (WebFetch could not parse it; read the binary directly). Fact 1's quote is verbatim on p.2 and establishes the regime, the named categories, and the named approval authority per category (Chairman of city/province People's Committee for establishment; DOET Director for operation of short-course training institutions). Fact 3's British Council quote is verbatim ('was passed by the Deputy Prime Minister in October 2024 and took effect in November 2024'). Fact 2's quote is verbatim on p.4 BUT DOES NOT ESTABLISH ITS CLAIM: the claim asserts 'an explicit capital threshold per institution type', while the source publishes thresholds ONLY for foreign-invested universities (VND1000bn) and branches (VND250bn) — the exact two categories the note itself concedes are 'plainly out of reach'. There is NO published capital threshold, and no published eligibility criterion of any kind, for the short-course training category on which the entire score rests. MAPPING FAILS on both limbs of anchor 3. Limb 1, 'a programme this project class demonstrably qualifies for': not demonstrated. The note asserts 'A Network School node is a foreign-invested education institution delivering short-course training' — that is the scorer's own characterization, not a sourced demonstration. The note concedes the test in its own words ('If the short-course-training category were shown not to cover a residential cohort programme... this drops to 2'), but the burden runs the other way: 'demonstrably qualifies' requires the scorer to SHOW fit, not the verifier to disprove it. A residential cohort community is not shown to be a short-course training provider. Limb 2, 'published eligibility criteria': for the relevant category, none were produced — only a named authority and a category label. Naming an approving official is not publishing eligibility criteria. SEPARATELY, THE INSTRUMENT IS THE WRONG KIND. Anchor 3's examples — SEZ regime, education-hub licence, digital-nomad statute — are opt-in facilitative instruments that signal receptivity. Decree 86 is mandatory baseline regulation that every foreign education investor must comply with; complying with a licensing law demonstrates nothing about a state's posture toward this project class. The axis is explicit: 'Demonstrated receptivity. Absence of hostility is not receptivity.' A generic regulatory permission regime is the paradigm case of absence-of-hostility. Anchor 2 fits precisely: 'a published investor-welcome or FDI policy not naming this project class' — Decree 86 is a published FDI-in-education policy naming only generic education categories, not this project class. The note's own downside analysis (IFC regime caps at 2 because it names financial services; SEZ law never enacted) is correct and reinforces that nothing lifts this above generic openness. REMAP to 2. Credit where due: the note's rejection of anchors 4 and 5 is correct and well-reasoned. ORIGINAL SCORING NOTE, retained — its facts stand and only the mapping moved: FACTS: Decree 86/2018/ND-CP, as amended by Decree 124/2024/ND-CP (in force 2024-11-20), is a published, rule-based licensing regime for foreign-invested education institutions. It names the institution categories (short-course training, schools, universities, branch campuses), names the approving authority for each step (Investment Registration Certificate → Establishment Licence → Operation Approval; provincial People's Committee Chairman for short-course training institutions), and publishes quantified capital thresholds. A Network School node is a foreign-invested education institution delivering short-course training, so it lands inside a named category with published criteria. MAPPING: anchor 3 — 'a programme this project class demonstrably qualifies for, published eligibility criteria (… education-hub licence …)'. WHAT ARGUED FOR HIGHER, AND WHY IT DID NOT REACH: anchor 4 requires a named empowered official or agency to have engaged with THIS project or a close analogue on the record. I searched for any Vietnamese government engagement with the Network School / network-state project class and found none — the Network School's only physical node is in Malaysia, and no Vietnamese official, ministry or People's Committee has been sourced saying anything about it or an analogue. Anchor 5 requires a signed MOU, gazetted incentive or allocated land naming the project or its class; nothing of the kind exists. Vietnam's IFC regime (Resolution 222/2025/QH15, in force 2025-09-01) offers real gazetted incentives including 10-year visas and work-permit exemption, but it names financial services in HCMC and Da Nang and does not name this project class, so on the rubric's own logic it caps at 2 and cannot lift the score. The SEZ route contributes nothing: the Law on Special Administrative-Economic Units was postponed indefinitely by the National Assembly on 2018-06-11 after mass protests and has never been enacted, so the instrument this class would most obviously fit does not exist. WHAT WOULD MOVE IT DOWN: anchor 0 is not met — no Vietnamese official or agency has been sourced opposing, restricting, revoking or investigating this project class. If the short-course-training category were shown not to cover a residential cohort programme (the branch-campus and university categories are plainly out of reach — top-500 global ranking, VND 250–1,000bn capital), the 'demonstrably qualifies' test would fail and this drops to 2 (generic FDI openness only). WHAT WOULD MOVE IT UP: any on-the-record statement from a named Vietnamese official or agency about this project or a close analogue → 4; a signed instrument → 5. | seed | seed | a513ea3e-f10d-47ba-8934-d616edc01ac3 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 174 | d88e5ada-4c32-4aa6-9fa5-06a658b3a1d9 | vn | visa-preference | 2 | rubric 0-5 | https://www.vietnam.vn/en/mien-thi-thuc-voi-chuyen-gia-nha-khoa-hoc-nha-dau-tu-nuoc-ngoai | Vietnam.vn (Vietnam national information portal), reproducing Báo Nhân Dân | 2025-08-15 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 2: A long-stay route exists but does not fit: wrong activity, wrong duration, or a cohort excluded. THE FACTS: · Vietnam's only preferential long-stay instrument in force is the Special Visa Exemption Card under Decree 221/2025/ND-CP, dated 2025-08-08, effective 2025-08-15. https://www.vietnam.vn/en/mien-thi-thuc-voi-chuyen-gia-nha-khoa-hoc-nha-dau-tu-nuoc-ngoai “The Government issued Decree No. 221/2025/ND-CP dated August 8, 2025, stipulating the temporary visa exemption for foreigners who are eligible for special preferential treatment to serve socio-economic development.” · The eligibility criteria are elite-only and exclude any ordinary Network School cohort: investors must be tied to a top-100 company by market capitalisation; scientists must hold a prestigious international award. https://www.vietnam.vn/en/mien-thi-thuc-voi-chuyen-gia-nha-khoa-hoc-nha-dau-tu-nuoc-ngoai “Regarding criteria, foreigners who are investors or leaders of large corporations and businesses worldwide must meet the following requirement: The corporation or business must be on the list of the 100 largest companies by market capitalization in the world, as published annually by reputable international organizations. Scientists in the field of science and technology must meet the following cr” · The route is not available as of right: every substantive category requires a written request from a ministry or Party/National Assembly agency, and the Government controls the list of bodies permitted to invite foreigners. https://www.vietnam.vn/en/mien-thi-thuc-voi-chuyen-gia-nha-khoa-hoc-nha-dau-tu-nuoc-ngoai “For cases stipulated in points b, c, d, and e above: There must be a written request from a ministry, ministerial-level agency, government agency, or Party or National Assembly agency.” · Duration per stay is capped at 90 days, so the card cannot deliver a 180-day continuous presence even for a qualifying holder. https://vietnamnews.vn/society/1721727/five-year-visa-exemption-considered-for-global-talents-including-top-scientists-executives-artists.html “Holders of the special visa exemption card would be granted 90 days of temporary residence each time they enter Việt Nam.” FACTS: Decree 221/2025/ND-CP (in force 2025-08-15) creates a Special Visa Exemption Card valid up to 5 years, multiple entry, 90 days of temporary residence per entry. Its published criteria are: guests of named Party/State leaders; scholars, professors, chief engineers and high-quality digital-technology personnel; investors/executives of companies in the world's top 100 by market capitalisation; award-winning scientists; top-100 footballers; honorary consuls. Every one of those categories additionally requires 'a written request from a ministry, ministerial-level agency, government agency, or Party or National Assembly agency', and the Government itself decides the list of institutes/universities/enterprises permitted to issue invitations. MAPPING: anchor 2 — 'A long-stay route exists but does not fit: wrong activity, wrong duration, or a cohort excluded.' It fails on all three limbs at once: wrong activity (it is a visa exemption for invited elites, not a work or study authorisation, and confers no right to conduct the project's activity class); wrong duration (90 days per entry against a 180-day requirement); and cohort fit (a Network School member is neither a top-100-company executive nor an internationally decorated scientist). WHAT ARGUED FOR HIGHER, AND WHY IT DID NOT REACH: anchor 3 requires a preferential route that FITS but is discretionary/capped/pilot-stage. Decree 221 is emphatically discretionary — which is the seductive part — but discretion is only the second half of anchor 3; the first half is fit, and it is not met, so 3 cannot be claimed. Anchor 4 requires a route open AS OF RIGHT on published criteria covering all four cohorts; the ministry-request requirement defeats 'as of right' outright. Anchor 5 requires a route designed for this activity class with published multi-year terms — Vietnam has no digital-nomad statute, and the 'Golden Visa' announced by the Prime Minister in spring 2025 has no decree, no criteria and no application portal as of retrievedAt, so per the rubric an intention is not a statute. WHAT WOULD MOVE IT DOWN: anchor 1 would require ordinary routes only; it does not apply, because Decree 221 and the DT investor visa series are genuine long-stay instruments that do exist. Anchor 0 does not apply: the ordinary route is the 90-day e-visa, which is open to citizens of all countries, so there is no named restriction or exclusion touching US/EU/IN/CN. WHAT WOULD MOVE IT UP: a remote-work or education route with published as-of-right criteria — the promised Golden Visa, if gazetted with criteria a member could actually satisfy → 3, or 4/5 once in force. SOURCING CAVEAT: the DT1/DT2 investor visas (5-year, VND 50–100bn+ capital thresholds) are a second long-stay route that also does not fit, but every source I found for their terms was a commercial visa agency, and the official texts were unreachable (vbpl.vn timed out; investdanang.gov.vn served an expired certificate), so I have not rested the score on them. The score stands on Decree 221 alone, which I verified on the Government's own portal. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS HOLD. Important process note: the vietnam.vn URL returned HTTP 403 to WebFetch, but this was a user-agent block, not a real barrier — curl with a browser UA returned HTTP 200 (992KB). I verified rather than degraded, because I could in fact check. All three vietnam.vn quotes are literally on the page and establish their claims: (1) 'The Government issued Decree No. 221/2025/ND-CP dated August 8, 2025, stipulating the temporary visa exemption for foreigners who are eligible for special preferential treatment to serve socio-economic development' — verbatim; (2) the top-100-market-cap requirement for investors and the 'prestigious and recognized international award' requirement for scientists — both verbatim; (3) 'For cases stipulated in points b, c, d, and e above: There must be a written request from a ministry, ministerial-level agency, government agency, or Party or National Assembly agency' — verbatim. MINOR SOURCING CAVEAT, NOT FATAL: fact 4's 90-day quote is literally on vietnamnews.vn, but that article describes the DRAFT decree ('a draft decree submitted by the Ministry of Public Security to the Ministry of Justice for review'), so as cited it evidences a draft term rather than the in-force one. The term did carry through — independent search confirms 'Foreigners using special visa exemption cards are granted a temporary residence certificate for 90 days for each entry' for enacted Decree 221, and this repo's own visa-access research in contracts/seed.sql independently records '5-yr card but only 90 days temporary residence per entry'. The vietnam.vn page itself confirms the 5-year cap and multiple entry. The 90-day limb is therefore substantively sound; it should be re-cited to an enacted-decree source. MAPPING HOLDS at anchor 2: 'A long-stay route exists but does not fit: wrong activity, wrong duration, or a cohort excluded.' A genuine long-stay instrument exists (5-year, multi-entry), and it fails all three limbs — wrong activity (a visa exemption for invited elites, conferring no work or study right), wrong duration (90 days/entry against 180), cohort excluded (no Network School member is a top-100-company executive or an internationally decorated scientist). The note resists the exact upward inference the rubric warns about: it correctly refuses anchor 3 on the ground that discretion is only anchor 3's SECOND half and fit is its first, and correctly refuses anchor 5 by treating the un-decreed 'Golden Visa' as an intention rather than a statute. Anchor 4 is defeated by the ministry-request requirement, which I verified verbatim. Anchors 0 and 1 correctly rejected. This score is honest and the sourcing caveat about DT1/DT2 is properly disclosed rather than hidden. | seed | seed | 8f2064f3-292f-4d70-9b96-3c66a5a0de54 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 175 | 282793c1-9691-4721-b916-755376bf82e2 | vn | web3-support | 2 | rubric 0-5 | https://theinvestor.vn/vietnam-considers-pilot-licenses-for-5-digital-asset-exchanges-d18648.html | The Investor (Vietnam Association of Foreign Invested Enterprises) | 2026-03-20 | 2026-07-17T00:00:00.000Z | REPUTABLE_MEDIA | CORROBORATED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 2: Legal but unregulated: no licensing regime; ramps operate informally. THE FACTS: · As of 2026-03-20, no crypto-asset exchange has been licensed in Vietnam. The Ministry of Finance has only shortlisted applicants; selection is still pending. https://theinvestor.vn/vietnam-considers-pilot-licenses-for-5-digital-asset-exchanges-d18648.html “the MoF will coordinate with the Ministry of Public Security and the State Bank of Vietnam to select up to five organizations” · A licensing framework exists on paper only: Resolution 05/2025/NQ-CP established a 5-year pilot with MoF licensing and a VND 10 trillion capital floor, but the pilot commences only once the first provider is licensed. https://english.luatvietnam.vn/legal-updates/government-resolves-to-facilitate-crypto-asset-market-operation-892-104216-article.html “After six months from the date the first organization providing crypto asset services is licensed by the Ministry of Finance, all crypto asset transactions by domestic investors must be conducted through this organization.” FACTS: Vietnam's Digital Technology Industry Law (National Assembly, June 2025) recognised digital assets as property, so crypto is legal. Resolution 05/2025/NQ-CP (2025-09-09) established a 5-year pilot licensing regime administered by the Ministry of Finance with a VND 10,000 billion (~USD 380–400m) charter-capital floor, 65% institutional ownership and a 49% foreign-ownership cap. Decision 96/QĐ-BTC opened applications on 2026-01-20. As of 2026-03-20 the MoF had received seven applications and had only shortlisted candidates — no licence has been granted, and the Vice Minister of Finance expects market activity to begin in Q3 2026. MAPPING: anchor 2 — crypto is legal but in practice unregulated, with no licensed venue operating and on/off ramps therefore still running informally (P2P). ANCHOR-FIT CAVEAT, STATED RATHER THAN HIDDEN: anchor 2's literal clause 'no licensing regime' is an imperfect match, because a licensing regime has been enacted; it simply has not issued a licence. Vietnam sits in a transitional state the rubric has no exact rung for. I map to 2 because it is the highest anchor whose substance (legal, no operating regulated venue, informal ramps) the facts actually satisfy. BANKING — the rubric's own note says anchors 1 and 3 both turn on banking, and DATA-SOURCES.md §194 lists banking access as having no automatable source. I could not source affirmative banking access to any primary source. The only banking instrument I found is SBV Directive 02/CT-NHNN (2018-04-13), which prohibits credit institutions from providing crypto-related payment, card, transfer and cross-border services; I could not source its repeal, but my sources for it were law-firm commentary, not the SBV itself, so I do not rest the score on it. This is the acknowledged weak point of this score. WHAT ARGUED FOR HIGHER, AND WHY IT DID NOT REACH: anchor 5's language — 'a named government programme, sandbox' — matches Resolution 05/2025/NQ-CP almost word for word, and this is precisely the upward inference the rubric forbids. The ladder is ordinal and cannot be skipped: anchor 3 requires that regulated exchanges OPERATE and that banking be obtainable, and neither conjunct holds as of retrievedAt — zero exchanges are licensed, and banking is unsourced. A sandbox that has licensed nobody is pilot-stage posture, not adoption; a zone's promotional posture is not receptivity, and by the same logic a pilot's promotional posture is not a functioning market. Anchor 4 additionally requires published tax treatment, which I did not find. WHAT WOULD MOVE IT UP: the MoF granting the first licence and that exchange trading with settlement through a named commercial bank → 3; published crypto tax treatment on top of that → 4. WHAT WOULD MOVE IT DOWN: sourcing that Vietnamese banks decline crypto service in practice would argue for 1, but 1's 'no framework' clause now fails on the Digital Technology Industry Law. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS HOLD. Both quotes re-fetched and verbatim. luatvietnam: 'After six months from the date the first organization providing crypto asset services is licensed by the Ministry of Finance, all crypto asset transactions by domestic investors must be conducted through this organization' — confirmed, and the page confirms it is Resolution 05/2025/NQ-CP, the VND 10 trillion charter-capital floor, and the five-year pilot. theinvestor.vn: 'the MoF will coordinate with the Ministry of Public Security and the State Bank of Vietnam to select up to five organizations' — confirmed, article dated 2026-03-20, and the page confirms selection is still pending with no licence granted. MAPPING HOLDS at anchor 2. I probed this hard for upward inference and found the opposite — the note is exemplary. It identifies the trap itself: anchor 5's language ('a named government programme, sandbox') does match Resolution 05 almost word for word, and the note refuses it because the ladder is ordinal and adoption has not occurred. That is exactly right; a sandbox that has licensed nobody is pilot-stage posture, not adoption. Anchor 3 is correctly refused because both conjuncts fail — zero exchanges operate (verified: selection pending as of 2026-03-20) and banking is unsourced. Anchor 4 is correctly refused for want of published tax treatment. Anchor 1 is correctly refused because its 'no framework' limb fails against the Digital Technology Industry Law. THE ACKNOWLEDGED IMPERFECTION IS REAL BUT DOES NOT MOVE THE SCORE: anchor 2's literal 'no licensing regime' is false — a regime is enacted, it has simply issued no licence. Vietnam sits between rungs the rubric does not have. Anchor 2 is nonetheless the correct landing spot, because it is the highest anchor whose SUBSTANCE (legal, no operating regulated venue, informal P2P ramps) the facts satisfy, and every higher anchor has a limb that definitively fails. Note also that the rubric's banking caveat cuts in this score's favour rather than against it: anchors 1 and 3 both turn on banking, which DATA-SOURCES.md §194 says is unsourceable — anchor 2 is the one rung that does not depend on the unsourceable element, so mapping here is not merely defensible but the only sourced option. Flagging the SBV Directive 02/CT-NHNN gap rather than resting on it was the correct call. | seed | seed | c4aa8709-e0c6-46b2-aca2-2ee0f13c06e1 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 176 | eeb06494-5549-4096-9cb5-d54bcea2e1ec | vn | capital-controls | 1 | rubric 0-5 | https://www.state.gov/reports/2024-investment-climate-statements/vietnam/ | U.S. Department of State | 2025-01-04 | 2026-07-17T00:00:00.000Z | OFFICIAL_SECONDARY | ANALYST_VERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 1: Reporting or registration only; no approval needed; no documented delays. THE FACTS: · There is no legal restriction on repatriating earnings or capital from Vietnam; the requirement is documentary, not an approval. https://www.state.gov/reports/2024-investment-climate-statements/vietnam/ “A foreign investor can convert and repatriate earnings, provided the investor has the supporting documents required by law, proving that they have completed financial obligations, including with respect to corporate income tax declarations and payments.” · The operative control is a notification obligation with a 7-working-day lead time — reporting, not approval. https://www.state.gov/reports/2024-investment-climate-statements/vietnam/ “SBV also requires foreign investors to submit notification of profit remittance abroad to tax authorities at least seven working days prior to the remittance.” · Remittance is legally permitted but carries a documentary burden; the reported friction is paperwork, not refusal, quota or delay. https://www.state.gov/reports/2024-investment-climate-statements/vietnam/ “Although foreign companies can remit profits legally, sometimes these companies encounter bureaucratic difficulties, as they are required to provide supporting documentation (e.g., audited financial statements, import/foreign-service procurement contracts, proof of tax obligation fulfillment, etc.).” · Controls do exist — domestic transactions must be in dong and remittance must route through authorised institutions — so this is not a zero. https://www.state.gov/reports/2024-investment-climate-statements/vietnam/ “Vietnam mandates that in-country transactions be made in Vietnamese dong. The government allows foreign businesses to remit lawful profits, capital contributions, and other legal investment earnings via authorized institutions that handle foreign currency transactions.” FACTS: Repatriation of profits and capital from Vietnam requires no approval. The controls are (a) registration — an FDI enterprise must maintain a Direct Investment Capital Account at a licensed bank; (b) reporting — notification to the managing tax authority at least 7 working days before remittance, per Circular 186/2010/TT-BTC; and (c) documentary proof that tax obligations are discharged. Domestic transactions must be denominated in dong, and remittance must route through authorised FX institutions. MAPPING: anchor 1 — 'Reporting or registration only; no approval needed; no documented delays.' All three limbs hold on the State Department's own language. SOURCE DISCIPLINE — REQUIRED DISCLOSURE: the rubric names IMF AREAER specifically, and I did not use it, because I could not reach it. elibrary-areaer.imf.org redirects every request to /Login/AuthServiceUnavailable.aspx (HTTP 302), and imf.org's AREAER publication page returns HTTP 403. Neither is dead — both are access-blocked, and I am reporting the block rather than working around it. This score is therefore reasoned from the U.S. State Department's Investment Climate Statement (OFFICIAL_SECONDARY, verified by direct fetch and quoted verbatim above) plus Circular 186/2010/TT-BTC as reported in secondary sources, NOT from AREAER. A verifier who reaches AREAER may legitimately land elsewhere. WHAT ARGUED FOR HIGHER (i.e. WORSE), AND WHY IT DID NOT REACH: two things. First, an earlier IMF characterisation surfaced in search describing 'limits on resident organisations' transfers and movements of USD to accounts abroad and restrictions on the purchase of FX for outward remittances' — that language points toward anchor 3 or 4, and I could not verify it against the underlying IMF document because of the 403/login wall, so I have not scored on it. It is the single biggest open risk on this score. Second, the State Department's 'bureaucratic difficulties' phrase invites a read of anchor 3's 'documented delays' — I rejected that read because the sentence describes a documentation burden, not a delay, a refusal or a discretionary approval, and I could not source an actual delay. Anchor 2 requires approval for large transfers; the State Department is explicit that the obligation is notification, so 2 does not reach. WHAT WOULD MOVE IT DOWN (i.e. BETTER): anchor 0 requires no controls at all; the DICA registration, the 7-working-day notification and the dong-only domestic rule are controls, so 0 is unavailable. WHAT WOULD MOVE IT UP: an AREAER entry, or a sourced instance of a blocked or delayed remittance, would move this to 3; the IMF's CFM language above, if verified, could reach 3–4. This score should be re-run against AREAER when access is available. CONFIRMED by independent adversarial verification (D-032/D-035): the facts were re-fetched at their sources and the mapping re-checked against contracts/rubric.md. VERIFIER: FACTS HOLD. Process note: WebFetch returned a 'Technical Difficulties' error page twice for state.gov; curl with a browser UA returned HTTP 200 (310KB), so I verified rather than degraded. All four quotes are literally on the page and establish their claims: (1) 'A foreign investor can convert and repatriate earnings, provided the investor has the supporting documents required by law, proving that they have completed financial obligations, including with respect to corporate income tax declarations and payments' — verbatim; (2) 'SBV also requires foreign investors to submit notification of profit remittance abroad to tax authorities at least seven working days prior to the remittance' — verbatim; (3) the 'bureaucratic difficulties'/supporting-documentation sentence — verbatim; (4) 'Vietnam mandates that in-country transactions be made in Vietnamese dong. The government allows foreign businesses to remit lawful profits, capital contributions, and other legal investment earnings via authorized institutions that handle foreign currency transactions' — verbatim. TWO SENTENCES THE NOTE DID NOT CITE STRENGTHEN IT: the section opens 'There are no legal restrictions on foreign investors converting and repatriating earnings or investment capital from Vietnam', and closes 'There are no recent changes or plans to change remittance policies that either tighten or relax access to foreign exchange for profit remittances.' I searched the retrieved text for any contrary language on delays, quantitative limits, or restrictions on FX purchase for outward remittance and found none. MAPPING HOLDS at anchor 1 — 'Reporting or registration only; no approval needed; no documented delays.' All three limbs hold on the source's own words: the obligation is notification (defeating anchor 2's 'approval required', correctly rejected), no approval exists, and no delay is documented. The note's rejection of anchor 3 is the right call and honestly reasoned: 'bureaucratic difficulties' denotes a documentation burden, not a delay, a refusal, or a discretionary approval — reading a delay into it would be inferring upward from friction to obstruction. Anchor 0 is correctly unavailable given DICA registration, the 7-day notification, and the dong-only rule. SOURCE DISCIPLINE SATISFIED: the rubric names AREAER and requires the note to say so if reasoning from elsewhere. The note discloses non-use, names the specific access failures (elibrary-areaer.imf.org 302 to a login wall; imf.org 403), reports the block rather than working around it, and flags the unverified IMF CFM language as the single biggest open risk instead of burying it. That is the disclosure the rubric asks for. The re-run-against-AREAER caveat should be carried forward. | seed | seed | 32df2851-0418-42ca-8c75-859e91f47a9b | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 177 | c3829d6f-493d-4e57-98d9-98f671665d5a | vn | policy-reversal-precedent | 3 | rubric 0-5 | https://vir.com.vn/fits-now-at-risk-through-retroactive-cuts-131718.html | Vietnam Investment Review | 2026-04-17 | 2026-07-17T00:00:00.000Z | REPUTABLE_MEDIA | UNVERIFIED | ANALYST_ENTRY | NULL | RUBRIC SCORE (D-035), a two-part claim. THE MAPPING: anchor 4: A programme closed or suspended to new entrants with little notice. THE FACTS: · THE INSTANCE: Vietnam is moving to revoke the feed-in tariff it previously granted to ~173 solar and wind projects, by retroactively applying a requirement introduced after those projects reached commercial operation. https://vir.com.vn/fits-now-at-risk-through-retroactive-cuts-131718.html “The above proposal is, in substance, retroactive, applying a newly introduced requirement (an acceptance certificate) to invalidate previously recognised COD dates and to revoke the FiT enjoyed by these projects. If implemented, more than 100 projects would lose their preferential tariff, and revenues would fall by 25-50 per cent.” · The reversal is directed by the state, not merely contemplated: the Government Inspectorate made the finding and the Ministry of Industry and Trade instructed EVN to recover or adjust the tariffs. https://vir.com.vn/fits-now-at-risk-through-retroactive-cuts-131718.html “The Government Inspectorate in 2023 noted that these projects continued to enjoy the preferential FiT despite the missing acceptance, thereby causing losses to EVN. The MoIT has asked EVN to review these projects, reassess their FiT eligibility, and recover or adjust tariffs for those that do not qualify, converting them to a lower purchase price.” · The incentive being reversed was a 20-year fixed tariff granted to attract the investment in the first place — i.e. a grant the state previously made and is now unwinding. https://vir.com.vn/fits-now-at-risk-through-retroactive-cuts-131718.html “During the 2017–2021 renewable energy boom, the government introduced feed-in tariff (FiT) mechanisms for solar and wind power to attract investment. Propelled by a 20-year fixed FiT, Vietnam swiftly became Southeast Asia's renewable-capacity leader.” · The reversal remains a proposal under negotiation as of April 2026, not a consummated revocation — which is what caps this below 5. https://solarquarter.com/2026/04/17/vietnams-evn-proposes-retroactive-tariff-cuts-for-renewable-projects-amid-13-billion-dispute/ “Vietnam's EVN Proposes Retroactive Tariff Cuts For Renewable Projects Amid $13 Billion Dispute” REMAPPED BY ADVERSARIAL VERIFICATION (D-032/D-035). The scorer claimed 4 at "anchor 4: A programme closed or suspended to new entrants with little notice.". An independent agent re-checked the facts against the pages AND the mapping against contracts/rubric.md, and the facts do not reach that anchor. VERIFIER: FACTS HOLD. All four quotes re-fetched and verbatim. VIR confirms the retroactive-acceptance-certificate mechanism, 'more than 100 projects would lose their preferential tariff, and revenues would fall by 25-50 per cent', the Government Inspectorate's 2023 finding, the MoIT instruction to EVN to 'recover or adjust tariffs', and the 2017-2021 boom with its 20-year fixed FiT. solarquarter confirms the headline and 17 April 2026 date, EVN's proposal of up to 43% reductions across 173 projects, and that 'a final decision from the government is still awaited'. MAPPING FAILS. Anchor 4 reads: 'A programme closed or suspended to NEW ENTRANTS with little notice.' The cited facts describe something categorically different — an unconsummated retroactive action against EXISTING holders. Neither limb of anchor 4 has cited support. I checked both sources specifically for the closure claim: solarquarter does not mention the FiT expiring or closing to new entrants, and VIR does not either. The note's supporting assertion that 'the FiT lapsed abruptly at the end of 2020 for solar and 2021 for wind, stranding transitional projects with no successor price mechanism' — the ONLY thing that would satisfy anchor 4's actual text — carries NO fact URL. Under the contract's own rule ('THE FACTS, each with a resolving source URL'), an unsourced assertion cannot carry an anchor. So anchor 4 rests on a fact that was never sourced plus a fact pattern that does not match its words. The note half-concedes this ('the fit is imperfect in a way I will not paper over'), then maps to 4 anyway. Note this axis is LOWER IS BETTER, so landing on 4 is the upward inference — it scores Vietnam WORSE on evidence that does not reach. CORRECTED TO 3. Walking the ladder against cited evidence only: anchor 5 ('grants revoked from existing holders') is correctly refused — a direction plus a proposal is not a revocation, and the note's own reasoning here is sound and well-disciplined. Anchor 2 fails because existing holders are precisely NOT grandfathered. Anchors 0 and 1 understate: there IS a reversal on record — a completed Government Inspectorate finding and a live ministerial instruction to recover or adjust granted tariffs — so this is not a case of absent history, and the note is right that UNRESOLVED is wrong. That leaves anchor 3: 'Terms tightened AND applied to existing holders, with a transition period.' This is the highest rung the cited facts actually reach — terms tightened and directed at existing holders via the MoIT instruction. The fit is imperfect in both directions and they roughly offset: the absent transition period argues worse, while non-consummation argues better. Per the contract's instruction to default REMAP-downward under anchor uncertainty, 3 is the honest landing. The note's own escalation trigger is the right one to carry forward: a sourced instance of EVN actually applying the reduced tariff to, or recovering payments from, a named existing PPA holder moves this to 5 — and would arguably justify skipping 4 entirely, since anchor 4's new-entrant framing never described this fact pattern. ORIGINAL SCORING NOTE, retained — its facts stand and only the mapping moved: FACTS: The instance is cited, as the rubric demands. Vietnam granted a 20-year fixed feed-in tariff to solar and wind projects during 2017–2021. In 2023 the Government Inspectorate found that ~173 projects (~USD 13bn of investment) had reached commercial operation without a construction completion acceptance certificate. The MoIT then instructed EVN to reassess eligibility and 'recover or adjust tariffs' — applying a requirement introduced in 2023 to invalidate COD dates recognised in 2017–2021. In April 2026 EVN formally proposed revenue reductions of up to 43% for pre-certificate generation, plus clawback of sums already paid, against holders of executed PPAs. Investors are threatening treaty arbitration. MAPPING: anchor 4. This is the honest rung, and the fit is imperfect in a way I will not paper over: the fact pattern is a directed-but-unconsummated retroactive revocation aimed at EXISTING holders, which is materially worse than anchor 3 (which requires a transition period — here there is none, the measure reaches backwards seven years) but not yet the completed act anchor 5 describes. Anchor 4's own text ('closed or suspended to new entrants with little notice') is separately satisfied on the same regime: the FiT lapsed abruptly at the end of 2020 for solar and 2021 for wind, stranding transitional projects with no successor price mechanism. WHAT ARGUED FOR HIGHER, AND WHY IT DID NOT REACH: anchor 5 — 'Grants revoked from existing holders' — is close, and a reasonable verifier could land there. I did not, because as of retrievedAt (2026-07-17) I could not source a completed revocation against a named existing holder. The MoIT has directed it and EVN has proposed it; developers are reportedly negotiating and some are accepting reduced terms rather than litigating. A direction plus a proposal is not yet a revocation, and scoring 5 on it would be inferring upward from an intention to an act — the exact move the contract forbids. WHAT WOULD MOVE IT UP: a sourced instance of EVN actually applying the reduced tariff to, or recovering payments from, a named existing PPA holder → 5. WHAT WOULD MOVE IT DOWN: if the measure were abandoned, or settled with a genuine transition period and no retroactive effect, → 3. Anchors 0–2 are unavailable: 0 and 1 require no reversal on record, and 2 requires existing holders to be grandfathered — the entire dispute is that they are not. Note the rubric's own warning cuts the other way here: this is not a case of absent history. Vietnam has a decade-old incentive regime and a documented record of moving against it, so UNRESOLVED is not the right answer. | seed | seed | 3196292e-e62b-4e95-98a7-8eef9d855636 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 178 | 27bee73b-9158-48f3-9a24-6693489d64c2 | my-jhr-forest-city-golf-hotel | availability | UNRESOLVED | NULL | https://fcgolfhotel.com/about-forest-city-phoenix-golf-hotel-johor-bahru | Country Garden Pacificview Sdn. Bhd. (operator) | 2026-07-17 | 2026-07-17T00:00:00.000Z | SELF_REPORTED | NEEDS_EVIDENCE | MODEL_ASSISTED_RESEARCH | NULL | AN ANALYST LOOKED AND THE QUESTION IS OPEN. The operator's home, about, facilities and conference-venue pages were searched for any statement about buyout, block booking, long-stay or group contracting, and none exists. Silence is not permission: the absence of a stated restriction establishes nothing. D-026 SPLITS THIS QUESTION AND THE SCHEMA DOES NOT: whole-property buyout and block booking are different instruments with different answers, and this gate has one field for both. No source addresses either. A booking engine showing free rooms was deliberately not consulted and would not have established availability if it had been — free inventory on a date is not a 36-month commitment. An availability date is obtainable only from the operator's sales desk (rsvgh_slcs@outlook.com, +60-135990209, per the Trip.com record), which is intake, not research (DATA-SOURCES.md:116). | seed | seed | bef35bd8-9fae-4037-8f80-4798016a01ac | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 179 | 4d08f699-a68d-4c4a-b632-994cfc788065 | ae-dubai-city | international-air-service | 1 | service (0/1) | https://ourairports.com/airports/OMDB/ | OurAirports (community open-data airport registry) | 2026-02-15 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | D-047 MIGRATION (2026-07-22): this row was seeded as `flight-connectivity` with value "HAS_SCHEDULED_INTERNATIONAL_AIRPORT" under unit "routes". Both the Almaty and the Tashkent verifiers independently flagged the shape: a string cannot be scored against a numeric unit, so the row was UNUSABLE in every city that carried it. D-044 licensed a degraded BINARY but never moved the criterion off `routes`. The owner's fix splits the two claims: `flight-connectivity` (route count) stays empty until OAG/Cirium is licensed, and the binary lands here as `international-air-service` = 1. The value below is unchanged in substance — only its subject, unit and encoding changed. Every quoted fact is the one the verifier re-fetched and CONFIRMED. OWNER RULING (D-044): OurAirports BINARY, degraded — no route count is available and that is the point of the ruling. VERBATIM from the OurAirports OMDB facility panel (fetch_page, 2026-07-18, re-fetched and CONFIRMED by the verifier): 'Name: Dubai International Airport', 'IATA code: DXB', 'ICAO code: OMDB', 'Facility type: large_airport', 'Airline service?: yes', 'Coordinates: 25.249790,55.370992', 'Last updated: 2026-02-15'. 'Airline service? yes' = CSV scheduled_service="yes"; 'Facility type: large_airport' = CSV type. A REAL route count needs OAG/Cirium (paid, not wired); OpenFlights routes are dead (DATA-SOURCES.md §6). TIER: OurAirports is a community-maintained public-domain registry, not a civil-aviation-authority primary (that would be the UAE GCAA/AIP). effectiveAt = the record's own 'Last updated' date. | seed | seed | ced0efd9-da55-4b1e-819a-50718db63cdb | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 180 | c126806d-35a7-49e6-ace8-34022bd93527 | ae-dubai-city | disaster-exposure | 2.3 | INFORM Risk 0-10 | https://data360api.worldbank.org/data360/data?DATABASE_ID=DRMKC_INFORM&REF_AREA=ARE | World Bank Data360 (relaying JRC/DRMKC INFORM Risk Index) | 2025-01-01 | 2026-07-18T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | OWNER RULING (D-044): INFORM as a COARSE SCREEN, tier-flagged, NATIONAL. VERBATIM from the World Bank Data360 keyless JSON API (fetch_page, 2026-07-18; re-fetched and CONFIRMED by the verifier): the record INDICATOR 'INFORM_OVRL', REF_AREA 'ARE', UNIT '0_TO_10', 'LATEST_DATA': true, 'TIME_PERIOD': '2025' carries OBS_VALUE '2.3' — a low-risk national score (compare Indonesia 4.6 for Bali). Same call returns 2025 sub-scores this composite launders: INFORM_HA (Hazard & Exposure) 4, INFORM_VU (Vulnerability) 1, INFORM_CC (Lack of Coping Capacity) 3. COARSE-SCREEN CAVEAT (DATA-SOURCES.md §5): INFORM 'is already a score' and launders its sub-scores — a COUNTRY-level composite used as a coarse screen, NOT a reviewable CITY observation. Attached to Dubai as INHERITED-NATIONAL. IMPORTANT: the low national 2.3 cannot see Dubai's real city-specific hazard — the documented US$2.9-3.4bn April-2024 record urban flood (see flood-sea-level) is exactly the signal a national roll-up flattens. TIER: the URL read is the World Bank Data360 mirror (MULTILATERAL) of the primary JRC/DRMKC INFORM index. effectiveAt = 2025 release year (start-of-year convention). | seed | seed | 56415dde-f644-4b0d-95ba-2f72d71ac80e | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 181 | a370b643-e26a-4ad5-815b-27db5922e401 | ae-dubai-city | flood-sea-level | 3 | rubric 0-5 | https://en.wikipedia.org/wiki/2024_United_Arab_Emirates_floods | Wikimedia Foundation (event article, citing National Center for Meteorology UAE, Guy Carpenter, CNN, Guardian, Reuters) | 2024-04-16 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | ANALYST_ENTRY | NULL | CONFIRMED by the adversarial verifier (all facts re-fetched verbatim; 3 judged the defensible centre — pure-elevation reading pulls toward 2, documented city-scale inundation pulls toward 4, and both correcting-down and inferring-up are barred). SCALE (contracts/rubric.md §8, Risk axis, LOWER is better; HIGHER rubric = MORE exposed). THE FACTS, each with a resolving URL: (1) Dubai is a coastal Persian Gulf city — VERBATIM https://en.wikipedia.org/wiki/Dubai: 'Dubai is on the Persian Gulf coast ... and roughly at sea level (16 m or 52 ft above).' The 16 m is a city-wide figure spanning inland desert; the built-up coastal districts, the Creek and ports sit far lower. (2) DOCUMENTED CATASTROPHIC URBAN INUNDATION — VERBATIM https://en.wikipedia.org/wiki/2024_United_Arab_Emirates_floods (fetch_page 2026-07-18): 'On 16 April 2024, heavy rains caused floods', 'the country's heaviest rainfall recorded in 75 years', 'At the Dubai airport, a total of 164 mm (6.45 in) of rain fell', 'A total of 1,244 flights at Dubai International Airport were cancelled', 'Five people were confirmed dead', damage 'US$2.9-3.4 billion'. ANCHOR MAPPING → 3: coastal built environment with defences that the April-2024 event demonstrably overwhelmed. WHY NOT 4: no 1-in-100-year flood-return map / coastal DEM / subsidence study in hand, and the one sourced city-elevation figure (16 m) sits above anchor 4's below-5 m threshold; the 2024 flood was PLUVIAL (extreme-rainfall/drainage), distinct from coastal sea-level inundation. WHY NOT lower: a coastal city with a documented deadly US$3bn+ record flood cannot sit at 'no coastal exposure'. UPSIDE-RISK FLAG for the property pass: a node on Palm Jumeirah, The World, or ANY reclaimed island is anchor 5 on its face (rubric.md §8: 'Reclaimed land is a 5 on its face'). This is the Dubai CITY-general call, not a site call. | seed | seed | 61280eae-27bd-45ea-bf80-97cda6047abc | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 182 | 71762959-992c-4c8e-b1ee-c3aa0a1207f7 | ae-dubai-city | local-additionality | 2 | rubric 0-5 | https://u.ae/en/about-the-uae/strategies-initiatives-and-awards/strategies-plans-and-visions/finance-and-economy/dubai-economic-agenda-d33 | The Official Platform of the UAE Government (u.ae), Dubai Economic Agenda D33 | 2023-01-04 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | CONFIRMED by the verifier; the cap-at-2 reasoning explicitly held as the inverse of Bali's 3. ANCHORS (contracts/rubric.md §2; D-035: 'economic weakness alone is not opportunity'; 3+ requires a PUBLISHED plan/statistic, else cap at 2). CRITICAL NUANCE — additionality is MARGINAL contribution, not absolute economic strength; a rich, saturated market caps the score even when plans exist. THE PUBLISHED PLAN — VERBATIM from u.ae (fetch_page 2026-07-18): the Dubai Economic Agenda D33 first-package projects include 'launching a programme to attract the world's best universities, making Dubai a global pioneering hub for higher education' — a published government plan naming higher education, which on its face points at anchor 3. BUT THE GAP IS SATURATED — VERBATIM https://en.wikipedia.org/wiki/Dubai_International_Academic_City: DIAC 'Consisting of 27 residential colleges and universities ... enrolls more than 27,500 students', 'more than 500 bachelor, master and doctoral programs', 'MENA's largest educational hub'; Dubai Knowledge Park has 'over 400 companies and institutions'. Dubai population ~4M, 92% expatriate. ANCHOR → 2 ('Neighbourhood-scale effect only, measurable within a district, invisible at city scale'): a ~300-person node is statistical noise in a 4M city and a marginal addition to a MENA-largest, already-dense higher-ed market — it fills NO unmet gap. WHY NOT 3: anchor 3 needs to fill a gap a published plan NAMES; D33 names higher-ed but DIAC/DKP already saturate it, so the node is substitutive/marginal — the D-035 point inverted (Dubai's saturation CAPS additionality, does not reward it). WHY NOT 1: D33 does name the activity, and a residential node has a district-measurable footprint. NEVER INFER UPWARD (rubric.md): a plan naming the sector does not lift a saturated market off the cap. effectiveAt = D33 approval date (2023-01-04). | seed | seed | 7e3ca8c2-212e-4e15-9b3b-1dd78837d2d6 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 183 | 1039fbc4-4b9e-4f07-9f2d-3ef04914def7 | ae-dubai-city | air-quality | NEEDS_EVIDENCE | PM2.5 ug/m3 | https://api.openaq.org/v3/sensors/10071461/years?limit=100 | AirGradient community sensor 'The Views', via OpenAQ v3 API | 2025-12-31 | 2026-07-22T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | OFFICIAL_API | NULL | HOLDBACK, not a wall: a fully verified candidate value EXISTS and is deliberately NOT asserted, because its deciding caveat needs an owner ruling the machine value cannot read. The 2026-07-18 key wall this row used to document is CLEARED (fetch-mcp.mjs authenticates to api.openaq.org; keyed pass + adversarial verification 2026-07-22, D-032/D-043: 7 CONFIRMED / 1 CORRECTED (immaterial) / 0 REFUTED — evidence/openaq-air/{dubai-claims,dubai-verdicts}.md). THE HELD VALUE, verbatim from the verifier's landing wording: LOW-COST SENSOR — owner ruling owed on grade acceptability. Annual-mean PM2.5 of 33.0 ug/m3 for calendar 2025 (local Asia/Dubai year window) from AirGradient community sensor 'The Views' (OpenAQ location 2981155, sensor 10071461), 16.2 km from the ae-dubai-city centroid; coverage 90% (7,889/8,760 hourly obs). Adversarially verified 2026-07-22 (independent re-fetch): value, coverage, distance, grade, licence, and monitor-absence all re-confirmed verbatim. Station selection: coverage tied at the API's reported 90.0% with Serena (7,898 h vs 7,889 h — a 9-hour margin); tie broken by distance to centroid (16.2 km vs 19.0 km). A strict raw-hours reading of 'best-covered' would instead select Serena at 40.9; same-grade complete-2025 siblings read 38.4 (Dubai Motor City) and 40.9 (Serena), so the honest same-grade range is 33-41 ug/m3. No reference/government PM2.5 monitor exists within the 25 km radius (monitor=true found=0; the API caps radius at 25000, 422 above) and the AE-wide enumeration (found:31, exhaustive) shows the only AE providers in OpenAQ are Abu Dhabi EAD (PM10-centric, stale since 2024-05-20), SPARTAN (Abu Dhabi, ~108 km away, stale since 2023-06-07), and AirGradient — no Dubai Municipality or consulate feed exists under any name. Licence CC BY 4.0 (OpenAQ licence id 41; commercialUseAllowed:true, attributionRequired:true; attribute 'Cheeseman'/AirGradient via OpenAQ). NOT a reference-grade or government value, NOT a city-wide average, NOT dust-corrected — desert dust inflates Dubai PM2.5 and the value is taken as measured. UPGRADE PATHS: (a) the owner rules low-cost grade acceptable -> land 33.0 as CORROBORATED with the note above; (b) a Dubai reference monitor appears in OpenAQ after 2026-07-22 -> it supersedes this candidate outright. Commercial mirrors (IQAir/AQICN) remain the WRONG source and licence — do not substitute them. | seed | seed | 85be88d1-f2f8-4379-ae6c-bce8bc9b439b | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 184 | d8ba19b5-3ad5-47a2-ac1d-01b6a4aac789 | id-bali-denpasar-city | international-air-service | 1 | service (0/1) | https://ourairports.com/airports/WADD/ | OurAirports (community open-data airport registry) | 2025-02-21 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | DOCUMENT_EXTRACTION | NULL | D-047 MIGRATION (2026-07-22): this row was seeded as `flight-connectivity` with value "HAS_SCHEDULED_INTERNATIONAL_AIRPORT" under unit "routes". Both the Almaty and the Tashkent verifiers independently flagged the shape: a string cannot be scored against a numeric unit, so the row was UNUSABLE in every city that carried it. D-044 licensed a degraded BINARY but never moved the criterion off `routes`. The owner's fix splits the two claims: `flight-connectivity` (route count) stays empty until OAG/Cirium is licensed, and the binary lands here as `international-air-service` = 1. The value below is unchanged in substance — only its subject, unit and encoding changed. Every quoted fact is the one the verifier re-fetched and CONFIRMED. OWNER RULING (findings/city-sources-bali.md): OurAirports BINARY, degraded — no route count is available and that is the point of the ruling. VERBATIM from the OurAirports WADD facility panel (fetch_page, 2026-07-18, re-fetched and CONFIRMED by the verifier): 'Name: Denpasar I Gusti Ngurah Rai International Airport', 'IATA code: DPS', 'ICAO code: WADD', 'Facility type: large_airport', 'Airline service?: yes', 'Coordinates: -8.748409,115.167123', 'Last updated: 2025-02-21'. 'Airline service? yes' = CSV scheduled_service="yes"; 'Facility type: large_airport' = CSV type. A REAL route count (routes to cohort hubs) needs OAG/Cirium (paid, not wired); OpenFlights routes are 12 years dead (DATA-SOURCES.md §6) and must not be used. TIER: OurAirports is a community-maintained public-domain registry, not a civil-aviation-authority primary (that would be Indonesia DGCA/AIP). effectiveAt = the record's own 'Last updated' date. | seed | seed | 208163e4-373b-442b-8694-5d054a1e710c | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 185 | d3c0b960-585f-4511-8d02-5939ba666e7a | id-bali-denpasar-city | disaster-exposure | 4.6 | INFORM Risk 0-10 | https://data360api.worldbank.org/data360/data?DATABASE_ID=DRMKC_INFORM&REF_AREA=IDN | World Bank Data360 (relaying JRC/DRMKC INFORM Risk Index) | 2025-01-01 | 2026-07-18T00:00:00.000Z | MULTILATERAL | CORROBORATED | OFFICIAL_API | NULL | OWNER RULING: INFORM as a COARSE SCREEN, tier-flagged, and it is NATIONAL. VERBATIM from the World Bank Data360 keyless JSON API (fetch_page, 2026-07-18; re-fetched and CONFIRMED by the verifier): the record INDICATOR 'INFORM_OVRL', REF_AREA 'IDN', UNIT '0_TO_10', 'LATEST_DATA': true, 'TIME_PERIOD': '2025' carries OBS_VALUE '4.6'. Same call returns the 2025 sub-scores this composite launders: INFORM_HA (Hazard & Exposure) 7.6, INFORM_VU (Vulnerability) 3.3, INFORM_CC (Lack of Coping Capacity) 4.0. COARSE-SCREEN CAVEAT (DATA-SOURCES.md §5): INFORM 'is already a score' and launders its sub-scores — this is a COUNTRY-level composite used as a coarse screen, NOT a reviewable CITY observation. Attached to Denpasar as INHERITED-NATIONAL; it says nothing Denpasar-specific (Bali's real profile is volcanic/seismic/coastal, which the national roll-up flattens). TIER: the URL read is the World Bank Data360 mirror (MULTILATERAL) of the primary JRC/DRMKC INFORM index; the DRMKC primary is a JS SPA + forced PDF download that fetch_page could NOT render on either the researcher's or the verifier's attempt (a same-instrument wall, D-043 — the world's fault, not a defect). effectiveAt = 2025 release year (start-of-year convention). | seed | seed | 203bd8a2-6dbb-4add-87ae-2f8a89138bc9 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 186 | 22b070ab-da4e-4414-a0e4-61b168704c00 | id-bali-denpasar-city | flood-sea-level | 3 | rubric 0-5 | https://ideas.repec.org/a/spr/nathaz/v120y2024i8d10.1007_s11069-024-06506-3.html | Natural Hazards (Springer); Widantara & Mutaqin, Universitas Gadjah Mada | 2024-06-01 | 2026-07-18T00:00:00.000Z | ACADEMIC_OR_INDEX | CORROBORATED | ANALYST_ENTRY | NULL | CORRECTED 4→3 by the adversarial verifier (anchor mapping, not a facts problem — every cited fact re-fetched and held). SCALE (contracts/rubric.md §8, Risk axis, LOWER is better; HIGHER rubric = MORE exposed). THE FACTS, each with a resolving URL: (1) Denpasar is coastal with a mean elevation ~4 m — VERBATIM https://en.wikipedia.org/wiki/Denpasar (fetch_page 2026-07-18): 'Denpasar is located at an elevation of 4 m (13 ft) above sea level.' Corroborated by OurAirports DPS field elevation 14 ft / 4 m MSL. (2) VERBATIM from the peer-reviewed Natural Hazards (2024) multi-hazard assessment of Denpasar (abstract on the IDEAS/RePEc mirror; DOI 10.1007/s11069-024-06506-3): 'gradual inundation, erosion, and coastal flooding have a low to very high level of hazard' and 'Denpasar's coastal region experienced an increase in shoreline by 4.2 km (8.7%) from 2012 to 2022'. VERIFIER'S ANCHOR CORRECTION: rubric.md §8 anchor 4 requires a 1-in-100-year flood-return map showing below-5m OR documented subsidence — neither instrument is in hand; the verified facts (coastal, ~4 m, a study reporting a RANGE 'low to very high', not uniform severity) land at anchor 3, not 4. NEEDS_EVIDENCE is NOT warranted (a hazard study exists → a positive score is right); a stricter anchor would demand the BNPB/BPBD flood-return map or a coastal DEM. UPSIDE-RISK FLAG for the property pass: the KEK Kura Kura SEZ site on Serangan Island is partly RECLAIMED land — a node there is anchor 5 on its face (rubric.md §8). This CITY score is the Denpasar-general call, not a site call. | seed | seed | c3bb88e5-feae-40c6-b2b8-b1ec6a87b9b3 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 187 | da423420-87a1-4c98-b2b9-dc434ee26a17 | id-bali-denpasar-city | local-additionality | 3 | rubric 0-5 | https://kek.go.id/id/investment/distribution/kek-kura-kura | Sekretariat Jenderal Dewan Nasional Kawasan Ekonomi Khusus (National Council for Special Economic Zones, Indonesia) | 2023-01-01 | 2026-07-18T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | ANALYST_ENTRY | NULL | ANCHORS (contracts/rubric.md §2; D-035: 'economic weakness alone is not opportunity'; 3+ requires a PUBLISHED plan or statistic, else cap at 2). CONFIRMED by the verifier via corroboration (the kek.go.id primary returned HTTP 200 but an empty JS-SPA body to BOTH agents — a same-instrument wall; the substance was corroborated out-of-band). THE PUBLISHED PLAN — VERBATIM from the SEZ authority page (fetch_page, waitForNetworkIdle, 2026-07-18): zone 'KEK Kura Kura Bali', location 'Pulau Serangan, Kota Denpasar, Provinsi Bali', legal basis 'PP No 23 Tahun 2023', area '498 ha', and the plan text naming 'taman pendidikan dan teknologi' (an EDUCATION-and-technology park) among sought activities. ANCHOR → 3: 'fills a gap a published local plan names' — a Network School (education) node fits the 'taman pendidikan dan teknologi' activity a gazetted government plan (PP 23/2023) seeks, in Denpasar Selatan; that clears the 3+ 'published plan' bar and lifts off the cap-at-2. WHY NOT 4/5: no sector statistic ranks a ~300-person node top-10 city-wide (anchor 4), no demand study for this activity, no municipal strategy naming Network School's class specifically (anchor 5). NEVER INFER UPWARD (rubric.md): the SEZ is primarily tourism/creative-industry/health; education is ONE named component, not the zone's purpose → 3, not higher. VERIFIER FLAG: the researcher's 'Target Investasi 89.9 Triliun' could not be re-verified (page walled; current sources cite Rp 104.4T) — not load-bearing for anchor 3. effectiveAt = year of PP 23/2023 (exact promulgation date not read — peraturan.bpk.go.id 403'd both fetchers). | seed | seed | 7bb8cb87-7c5c-44b5-9e3a-904bc820fe3a | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 188 | 1e959d7c-0c00-4c7c-9d02-3418fdc06d96 | id-bali-denpasar-city | air-quality | NEEDS_EVIDENCE | PM2.5 ug/m3 | https://api.openaq.org/v3/locations?coordinates=-8.6717,115.2339&radius=25000&limit=100 | OpenAQ | 2026-07-22 | 2026-07-22T00:00:00.000Z | OFFICIAL_SECONDARY | NEEDS_EVIDENCE | OFFICIAL_API | NULL | DATA-COVERAGE HOLD, not an access wall and not a world-absence claim — the 2026-07-18 key wall this row used to document is CLEARED. The keyed pass ran 2026-07-22 (fetch-mcp.mjs now authenticates to api.openaq.org; every fetch showed 'authenticated with OPENAQ_API_KEY'), researcher + adversarial verifier per D-032/D-043 (evidence/openaq-air/{bali-claims,bali-verdicts}.md, 7 CONFIRMED / 1 CORRECTED / 0 REFUTED). WHAT THE KEYED API SHOWS: ZERO reference-grade monitors on Bali — 'monitor=true' returns found:0 within 25 km AND across an all-Bali bbox (incl. Singaraja/Kintamani); iso=ID&monitor=true lists 8 Indonesian reference monitors, all Jakarta/Bandung; no KLHK/ISPU or consulate feed for Bali exists in OpenAQ under any name. All 5 Bali stations are AirGradient low-cost sensors, and none has a usable complete year: the best record is Ubud (location 4826556, 16.1 km away, a different airshed) at 17.7 ug/m3 for 2025 but only 45% coverage (Jun 22-Dec 5, dry-season-biased, silent since); Balangan 7%; Kuwum/Umadawa/Tabanan live only since June 2026 with no yearly rollups. The 17.7 is a fetch fact, NEVER a proposable value. LICENCE, for whenever a value does exist: CC BY 4.0, verbatim 'commercialUseAllowed":true' (https://api.openaq.org/v3/licenses/41) — the licence is not the blocker. RE-CHECK EARLY 2027 (the closest path to a value): Kuwum /v3/sensors/16625235/years (8.0 km from centroid, live since 2026-06-15), Umadawa 16632095, Tabanan 16694904 (the claims file's 16694905 is the RH sensor — corrected by the verifier). Any future proposal from these is LOW-COST SENSOR — owner ruling owed on grade acceptability. API behaviour recorded: radius caps at 25000 (422 above), so widening is done by bbox. Commercial mirrors (IQAir/AQICN) remain the WRONG source and licence — do not substitute them. | seed | seed | 6157aa82-54c2-4858-8653-bd176cf7a0ed | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 189 | 9e57dfc3-9616-4cd1-a59a-14ae23e81f75 | kz-almaty-city | air-quality | 24.3 | PM2.5 ug/m3 | https://api.openaq.org/v3/sensors/25903/years?limit=100 | AirNow (US government air-quality data system; OpenAQ owner field: 'Unknown Governmental Organization'), via OpenAQ v3 API | 2024-12-31 | 2026-07-22T00:00:00.000Z | OFFICIAL_SECONDARY | CORROBORATED | OFFICIAL_API | NULL | THE KEYED OPENAQ PASS, 2026-07-22 — the row four agents across four cities could not write while the fetcher sent no key (the 401 wall was ours, not the world's; fetch-mcp.mjs now authenticates to api.openaq.org by exact hostname). LICENCE: US Public Domain, verified in the API's own licence record, verbatim: '{"id":33,"name":"US Public Domain","commercialUseAllowed":true,"attributionRequired":false,"shareAlikeRequired":false,"modificationAllowed":true,"redistributionAllowed":true,"sourceUrl":"http://www.usa.gov/publicdomain/label/1.0/"}' (https://api.openaq.org/v3/licenses/33). Commercial use allowed; no attribution required — the first air-quality row in the portfolio with an unrestricted licence. MONITOR GRADE: reference-grade government monitor — location record verbatim: 'isMonitor":true', 'instruments":[{"id":2,"name":"Government Monitor"}]', provider AirNow (id 119), owner 'Unknown Governmental Organization' (id 4) — the ONLY reference monitor within 25 km of the kz-almaty-city centroid ('found":1' on the monitor=true query), 3708 m from it, at (43.252855, 76.93118). WHAT THIS VALUE IS: the OpenAQ v3 yearly rollup of hourly PM2.5 at that single monitor for calendar year 2024 (period 2024-01-01T00:00+06 to 2025-01-01T00:00+05 local), API-stated value 24.3, underlying avg 24.252346570397112, coverage 8319 of 8784 expected hourly observations = 94.7% (API-stated 95.0). WHAT IT IS NOT: not a city-wide or population-weighted mean (AAI 2024 districts ran 16.1 south to 31.2 north-east); not a current-year figure (the monitor's datetimeLast is 2025-11-14T11:00Z, so 2025 is incomplete and its partial rollup of 25.5 is NOT this value); not seasonal (Almaty winter 2024 peaked ~49.9 vs summer 11.5 per AAI — an annual mean hides a ~5x inversion swing). DATA-QUALITY FLAGS, EXPANDED BY THE VERIFIER: the 2024 rollup carries 'hasFlags":true'; /v3/locations/8876/flags returns 54 records, all 'Limits exceeded' (small negative hourlies, -1..-3 ug/m3, an instrument-near-detection-limit behaviour); 8 records spanning ~9 hours fall inside 2024; effect on the annual mean < 0.01 ug/m3 — benign. CORROBORATION, context only: AAI's independent 2024 Clarity-network mean was 24.08 ug/m3; AAI's tabulation of Kazhydromet's 2024 average was 20; a long-running AirGradient low-cost sensor (OpenAQ sensor 11235891) shows 36.4 for complete-2025 at 90% coverage — low-cost optical, uncorrected, single site, recorded as prose context never as the value. DISCREPANCY: supersedes the REFUTED World Bank/IIASA modelled 60 (2018); 24.3 sits inside the deep-pass verifier's district-bound band [16.1, 31.2] and confirms the overturn. FEED-DARK CAUTION: the monitor has been dark since 2025-11-14; CREA's 4-Mar-2025 AirNow-halt report does not explain it (data flowed for months after March); cause unestablished — the 2026 re-check should first ask whether the feed resumed. VERIFIER: CONFIRMED, 15 CONFIRMED / 0 CORRECTED / 0 REFUTED / 0 UNVERIFIED — every fragment re-fetched independently, all 200-with-AUTH (evidence/openaq-air/{almaty-claims,almaty-verdicts}.md). | seed | seed | d6989d7c-c6cc-459b-9609-4404279e0739 | NULL | 2026-07-22T07:56:50.788Z | NULL |
| 190 | 5d732290-818b-4ef4-8012-45635f3a5b06 | kz-almaty-city | sunshine-hours | 2278.2 | hours/yr | https://www.nodc.noaa.gov/archive/arc0216/0253808/3.3/data/0-data/Region-2-WMO-Normals-9120/Kazakhstan/CSV/ALMATYOGMS_36870.csv | NOAA NCEI, archiving the WMO 1991-2020 Climatological Standard Normals (station ALMATY OGMS, WMO 36870, submitted by Kazhydromet) | 2020-12-31 | 2026-07-22T00:00:00.000Z | OFFICIAL_PRIMARY | CORROBORATED | DOCUMENT_EXTRACTION | NULL | RETRIEVAL PATH — DISCLOSED, PER D-050, BECAUSE OUR OWN INSTRUMENT CANNOT RENDER THIS URL. The NOAA archive serves this CSV as `application/octet-stream`, and Chromium downloads octet-streams, so `fetch_page` fails with 'page.goto: Download is starting' on every NOAA host, both schemes, and on the XLS mirror. Fourteen non-proxy routes were tried across two agents and none rendered the contents; Wayback has NO capture of the file (404), which is the route that would solve this citably. The sheet was read by fetching THIS SAME NOAA URL through the `r.jina.ai` text-rendering proxy, independently by the researcher and again by the adversarial verifier. THE PROXY IS NOT BYTE-TRANSPARENT and the note must not claim otherwise: it normalises CRLF to LF (6973 chars returned + the sheet's 126 line endings = 7099 = 6.93K, matching NOAA's own 6.9K listing). Two NOAA hosts through ONE proxy tests host agreement, not the return path. WHAT ACTUALLY CARRIES FIDELITY, and it is strong: the twelve monthly values sum to 2278.2 in exact decimal (recomputed independently twice), and the Wikipedia climate box — a completely separate retrieval, years earlier, by a different agent — carries the identical twelve months and identical annual total digit for digit. That is an independent witness, not a superseded source. VERBATIM from the sheet: '36870,8,Sum,4,112.0,119.0,163.0,187.8,235.3,260.0,280.3,277.7,243.6,188.9,116.6,94.0,2278.2' under the parameter header '8,Total_Number_of_Hours_of_Sunshine,hours'. COMPLETENESS: the sheet's own NOY row shows 30 years present in all twelve months — a full normal, not gap-filled. STATION SITING — THE FORMER CAVEAT IS RESOLVED AND STRUCK, on sources that need no proxy at all. The sheet gives '36870,43|14|22|N,76|55|58|E,846.870000'; WMO OSCAR/Surface independently returns latitude 43.239444, longitude 76.932778, elevation 846.87, GSN member, established 1859, and Kazhydromet (the submitting NMHS) and WWIS both render directly. That is 3.49 km from the seeded centroid [76.8897, 43.2389] on a bearing of 089 degrees — same latitude to within ~60 m, on the valley floor at the low end of the city's 500-1700 m range, inside the built-up area. It is a city-centre station, NOT a mountain-slope one, which is the direction the inversion-basin worry pointed. An apparent elevation conflict (OSCAR 846.87 m vs Kazhydromet 851 m) is OSCAR's own barometer height, not a discrepancy. And the release's other Almaty station, Kamenskoe Plato (36875, 1316 m, the upper city), publishes NO sunshine parameter — so 36870 is the only Almaty sunshine normal in the primary and there is no station-choice ambiguity. NO COMPETING FIGURE IS RECORDED: commercial climate sites show a wild spread (~2280 to ~3774), none was retrieved, and a search summary is not a retrieval. D-049 DATING: effectiveAt = 2020-12-31, the END of the 1991-2020 normal period. | seed | seed | 7c328830-b454-4bc2-94ae-9da75627afea | NULL | 2026-07-22T07:56:50.788Z | NULL |