OFAC Sanctions Screening API (SDN, UK OFSI, UN, EU Lists)

Screen customers, suppliers and counterparties against the OFAC, UK OFSI, UN and EU sanctions lists with explainable fuzzy matching, for compliance teams, finance ops, KYC onboarding and AI agents. It costs $0.01 per name screened, however many lists you pick or hits come back, so 100 names cost $1. It is a screening aid, not legal advice.

Screen customers, suppliers, counterparties, vessels and beneficial owners against the official sanctions lists, downloaded fresh from the issuing authorities on every run: the US Treasury OFAC SDN list, the OFAC Consolidated (non-SDN) list, the UK OFSI consolidated list, the UN Security Council consolidated list and, optionally, the EU consolidated financial sanctions list. Matching is fuzzy and explainable: every hit carries a 0 to 100 score, the matched primary name or alias, entity type, programs or regime, countries, listing date and the authority's own remarks.

Built and maintained by Lathewright. Lathewright is operated by AI systems under human ownership. A person reviews and approves everything we publish, send and sell. $0.01 per name screened, regardless of how many lists you select or how many hits come back.

Who uses it

• KYC/KYB onboarding: screen new customers and companies before activation; store the row as the audit record.
• Accounts payable and procurement: batch-screen vendor masters monthly.
• Marketplaces and fintech agents: an LLM agent calls this tool, reads decision and topScore, and escalates potential-match rows to a human.
• Investigations: see aliases and programs across all lists in one place.

Example tasks

Paste any of these into the Input tab, or send the same JSON through the API. Prices are the per-result prices listed under Pricing.

1. Screen three names against the default lists

{
  "names": ["Rosneft", "Viktor Bout", "Acme Widgets LLC"]
}

What comes back: One row per name with decision, topScore, matchCount, listsScreened and a matches list with the matched name or alias, programs and listing date. $0.03.

2. Stricter threshold, chosen lists

{
  "names": ["Rosneft"],
  "lists": ["ofac-sdn", "eu"],
  "minScore": 90,
  "maxMatchesPerName": 3
}

What comes back: Only matches scoring 90 or higher on the OFAC SDN and EU lists, at most 3 per name.

3. Limit matches to vessels and organizations

{
  "names": ["Example Shipping Co"],
  "entityTypes": ["vessel", "entity"]
}

What comes back: Individuals and aircraft are ignored when matching.

Input

{
  "names": ["Rosneft", "Viktor Bout", "Acme Widgets LLC"],
  "lists": ["ofac-sdn", "ofac-consolidated", "uk-ofsi", "un"],
  "minScore": 85,
  "maxMatchesPerName": 5,
  "entityTypes": []
}

Field	Type	Notes
names	array	People, companies, vessels (one per line)
lists	array	ofac-sdn, ofac-consolidated, uk-ofsi, un, eu (EU adds ~25 MB and ~20 s)
minScore	50 to 100	Similarity threshold; 85 default, 75 wide net, 95 near-exact
maxMatchesPerName	integer	Best hits to return per name
entityTypes	array	individual, entity, vessel, aircraft

Output (one row per screened name)

{
  "name": "Viktor Bout",
  "decision": "potential-match",
  "topScore": 90,
  "matchCount": 1,
  "listsScreened": ["ofac-sdn", "ofac-consolidated", "uk-ofsi", "un"],
  "matches": [
    {
      "score": 90,
      "list": "ofac-sdn",
      "entityId": "9272",
      "matchedName": "BOUT, Viktor Anatolijevitch",
      "matchedNameType": "primary",
      "entityType": "individual",
      "programs": ["DRCONGO"],
      "countries": [],
      "listedOn": null,
      "details": "DOB 13 Jan 1967; alt. DOB 13 Jan 1970; POB Dushanbe, Tajikistan; ...",
      "sourceUrl": "https://ofac.treasury.gov/specially-designated-nationals-and-blocked-persons-list-sdn-human-readable-lists"
    }
  ],
  "minScore": 85,
  "listVersions": { "ofac-sdn": { "records": 39653, "ms": 3162 }, "uk-ofsi": { "records": 19761, "version": "03/06/2026" }, "un": { "records": 3776, "version": "2026-10-17T23:00:01.054Z" } },
  "screenedAt": "2026-10-19T05:12:40.000Z"
}

How scoring works: names are normalized (accents, punctuation, case, corporate suffixes), then compared on sorted tokens (Jaro-Winkler), token overlap and per-word alignment, so Victor But finds BOUT, Viktor and Rosneft finds all three ROSNEFT entries. Generic words (technologies, bank, international, trading…) are down-weighted so Apify Technologies does not match Poly Technologies. Aliases (a.k.a., f.k.a.) are searched as well as primary names.

Pricing

Event	Price
Name screened (name-screened)	$0.01

Plus Apify's start fee ($0.00005). 1,000 names = $10. The lists are cached for the day in a key-value store named sanctions-lists-cache in your account, so repeated runs on the same day skip the downloads.

How to use it

• Console: paste names, Start, export the rows as your audit trail.
• API: POST https://api.apify.com/v2/acts/lathewright~sanctions-screening/run-sync-get-dataset-items?token=<token>
• n8n / Make / Zapier: Apify node → screen each new CRM contact or vendor → route potential-match to a review queue.

Use from an AI agent (MCP)

Add this Actor as a tool in Claude, Cursor, VS Code or any MCP client with https://mcp.apify.com/?tools=lathewright/sanctions-screening. The Actor description is the tool description, and the input fields above are the tool parameters. Example prompts: "Screen these five vendors against OFAC and UK sanctions and tell me which need review", "Is Rosneft on any sanctions list and under which programs?".

Data sources, fair use and limits

• OFAC SDN and Consolidated lists: Sanctions List Service CSV exports (public domain).
• UK OFSI consolidated list: HM Treasury CSV (Open Government Licence).
• UN Security Council consolidated list: official XML.
• EU consolidated list: European Commission CSV (reuse permitted).
• Screening a name is a suggestion, not a legal determination. Treat potential-match as "review by a person", confirm with identifiers (date of birth, registration numbers, addresses) and keep the row as evidence. PEP (politically exposed persons) and adverse media are not included.

FAQ

Why does a common name return several matches? Fuzzy matching surfaces similar names on purpose. Raise minScore, restrict entityTypes, or compare the details field (dates of birth, places) with what you know.

How fresh are the lists? They are downloaded from the authorities at run time (cached for the current day). listVersions shows record counts and publication stamps where the source provides one.