Sanctions Screening — OFAC & UN Watchlists

Screen people and companies against the official OFAC SDN, OFAC Consolidated and UN Security Council sanctions lists. Alias-aware fuzzy matching with a 0–100 score and risk level, so spelling variants like AL-QA'IDA / al qaida and Ivan Ivanov / IVANOV, Ivan Ivanovich are actually caught.

Free official government sources. No API key, no login, no anti-bot.

Why this one

The lists are free — the friction was never access. It's that OFAC ships a headerless, fixed-column CSV with aliases in a separate file, the UN ships a bespoke XML, and screening a name needs alias-aware fuzzy matching, not string equality. This actor does all three and hands you a decision.

Two modes

• Screen — give it names → get candidate matches with a score, risk level, sanctions programs and source.
• Dump — leave names empty → get the consolidated, deduped list itself as a dataset.

What you get

Field	Description
query	The name you screened
matchScore	0–100 similarity (100 = exact after normalisation)
riskLevel	confirmed_match · high · medium · low · no_match
matchType	exact · exact_alias · name · alias
matchedName / matchedOn	The listed entry, and the exact string that matched
entityType	individual · entity · vessel
programs	Sanctions programs (e.g. UKRAINE-EO13662, Al-Qaida)
aliases	Known aliases for the listed entry
title / remarks	Designation and notes (DOB, passport, addresses)
sourceLabel / uid	Which official list, and its record ID
status	match or clear (clean names are returned too, not dropped)

How to use it

1. Paste the names to screen (one per line is fine).
2. Pick the lists — OFAC SDN alone is the common default; add UN Consolidated for broader coverage.
3. Tune minimum match score: 75 balanced · 60 catches more spelling variants (more false positives) · 90 near-exact only.

Input

{
  "names": ["Ivan Ivanov", "Acme Trading LLC"],
  "sources": ["ofac_sdn", "un_consolidated"],
  "minMatchScore": 75,
  "maxMatchesPerName": 10
}

Sample output

[
  {
    "query": "Ivan Ivanovich Ivanov",
    "matchScore": 100,
    "riskLevel": "confirmed_match",
    "matchType": "exact",
    "matchedName": "IVANOV, Ivan Ivanovich",
    "matchedOn": "IVANOV, Ivan Ivanovich",
    "entityType": "individual",
    "programs": "UKRAINE-EO13662",
    "aliases": "IVANOV, Vanya | I. Ivanov",
    "remarks": "DOB 01 Jan 1970; nationality Russia",
    "sourceLabel": "OFAC SDN (US Treasury)",
    "uid": "12345",
    "status": "match"
  },
  {
    "query": "Acme Trading LLC",
    "matchScore": 0,
    "riskLevel": "no_match",
    "matchedName": null,
    "status": "clear"
  }
]

Typical uses

• KYC / onboarding checks — screen a customer or supplier list before you transact.
• Continuous monitoring — schedule it over your customer book; lists change and yesterday's clear name may be listed today.
• Vendor & counterparty due diligence — batch-screen a procurement list.
• Building a compliance dataset — dump mode gives you the consolidated list to load into your own system.

Pricing

$2.00 / 1,000 screened names ($0.002 per result), plus a near-zero start fee. Clean names are returned as clear rows so you have an auditable record of what was checked. Never charged beyond maxItems.

⚠️ Important — what this is and isn't

• This is decision-support, not legal advice, and not a certified compliance determination. A match is a candidate for human review, never an automatic conclusion.
• Fuzzy matching cuts both ways: a low threshold produces false positives; a high one can miss transliteration variants. Always have a human review hits before acting.
• Coverage is limited to the lists you select. It does not include every national list (e.g. EU, UK OFSI, Swiss SECO) — absence of a match is not proof a party is unsanctioned.
• Lists are fetched live at run time, so results reflect the sources as of that run. Record the run date.

FAQ & limitations

• Do I need an API key? No. OFAC and the UN publish these files openly.
• Are aliases included? Yes — OFAC's separate ALT.CSV is fetched and attached, so alias hits are found.
• Why did a clean name still return a row? Deliberate: a clear row proves the name was screened.
• How current is the data? As current as the source files at run time; OFAC updates frequently.
• Can I add more lists? EU/UK lists aren't wired yet — the parser is source-agnostic, so they're a natural next addition.

Related actors

• Email Verifier · Website Contact Finder — the rest of a counterparty-diligence stack.
• SEC EDGAR Filings — corporate filings and material events for the same entities.
• GitHub Repo Scraper · npm Package Info — supply-chain diligence on the technical side.