OFAC SDN Fuzzy Name Matching - $0.02/Name, No API Key
Pricing
from $3.20 / 1,000 sanctions records
Screen names against the official U.S. Treasury OFAC SDN and Consolidated sanctions lists for $0.02 per name. Fuzzy matching (typos, aliases), CSV batch input, downloadable audit trail. No API key, no monthly subscription.
Individual or entity names to screen against the sanctions lists. By default (matchMode=fuzzy) this checks the official SDN_Name field, OFAC's own a.k.a./f.k.a. aliases, and a typo/reorder-tolerant fuzzy score. Leave empty (and namesCsv empty too) to fetch the full list instead (subject to maxRecordsPerRun). Each entry can be a plain string, OR an object {"name": "...", "dateOfBirth": "YYYY" or "YYYY-MM-DD", "country": "..."} to supply secondary identifiers that disambiguate a common-name hit from an unrelated same-named real designee (see README "Secondary identifiers") - dateOfBirth/country are optional even in object form. Example: ["Example Trading Co", {"name": "Vladimir Putin", "dateOfBirth": "1952", "country": "Russia"}].
[]Paste CSV text as an alternative (or addition) to the names list above - useful for a batch of many names at once. Optional header row: a "name" column (required if you use a header at all) and optional "referenceId" (your own case/tracking ID), "dateOfBirth" ("YYYY" or "YYYY-MM-DD"), and "country" columns - the last two supply secondary identifiers, same as the names array's object form (see README "Secondary identifiers"). Without a header, the first column of every line is treated as the name. IMPORTANT: if a name itself contains a comma (e.g. OFAC's own "Last, First" style, "Smith, John"), quote it: "Smith, John",CASE-1 - an unquoted comma is parsed as a column separator like any standard CSV. Combined with names if both are set.
Which official Treasury lists to check. SDN = Specially Designated Nationals (primary sanctions list). CONSOLIDATED = the non-SDN consolidated sanctions list (e.g. Foreign Sanctions Evaders, Sectoral Sanctions Identifications). Any REAL screen (names provided) always defaults to BOTH lists regardless of this field's own default — SDN coverage is never silently dropped from an actual screen. Only a run with no names to screen at all uses a cheaper single-list default; that run always discloses exactly which list(s) were and weren't included via a lists-coverage output record. Example: ["SDN", "CONSOLIDATED"] (the default — most callers should leave this alone).
[ "SDN", "CONSOLIDATED"]Optional: restrict results to specific sanctions programs (e.g. "CUBA", "IRAN", "DPRK3"). Leave empty for no filter. Program codes are as published by OFAC; check a full-list run's output for the exact codes in use if unsure.
[]When on, the actor remembers which entity numbers (ent_num) it has seen before (via Apify key-value store) and emits an additional 'new-listing' record for anything that's newly appeared on the list since the last run. Designed for scheduled runs watching for new sanctions designations. Charges one monitor-run PPE event in addition to per-record events. Combine with watchlistName (below) for a different, name-scoped alerting mode: alert only when one of YOUR screened names newly becomes a match, not every new list-wide designation.
Set this (any label of your choosing, e.g. "Vendor List Q3") together with monitorMode=true and a real screen (names/namesCsv) to turn this actor into a recurring change monitor for YOUR OWN list: set up an Apify Schedule with the SAME input every time, and each run will push a watchlist-alert record only for a (name, list, entity) match that is NEW since this exact watchlist's last run - not a repeat of an already-known match, and not every new designation on the full OFAC list (that's the separate, list-wide monitorMode behavior above). Different watchlistName values are tracked completely independently, even within the same Apify account - safe to run several distinct customer lists through this one actor. Leave empty for the original (pre-2026-09-24) monitorMode behavior only. No extra charge - reuses the existing monitor-run PPE event.
Applied INDEPENDENTLY to each list in lists (not shared across them). Default 0 = no cap — every matching record on every requested list is returned, because a sanctions screen must be exhaustive by default; a silently-partial screen is the worst failure mode for this actor. If you set a cap and it truncates a list, the actor emits a loud screening-status TRUNCATED record and a warning log — it never reports a capped run as a complete screen. The full SDN + Consolidated lists together are roughly 17,000-20,000 entries. IMPORTANT: if you request a full-list export with no names (e.g. explicitly setting lists: ["SDN"] with nothing to screen) and leave this at 0, fetching and pushing the entire ~19,000-row SDN list can exceed Apify's own per-run cost limit and abort partway through — set an explicit cap for full-list exports.
"fuzzy" (default): checks each name against the primary SDN_Name field, OFAC's own a.k.a./f.k.a. aliases (both the structured alias export AND aliases embedded in the remarks field), using exact matching, common transliteration folding (c/k, v/w, y/i/j, and more), and a one-to-one per-token typo-tolerant score (see fuzzyThreshold). Live-benchmarked recall 97.6%, false-hit rate 3.5% (see README). This is additive over the old behavior - every match "exact" mode would have found, it still finds, plus new alias/transliteration/near-miss matches. "exact": restores the original, literal case-insensitive substring-only matching against SDN_Name alone (no aliases, no fuzzy scoring) - use this if you need output identical to this actor's pre-2026-09-24 behavior.
Minimum rapidfuzz-style token-sort-ratio score (0=totally different, 100=identical after normalization) for a fuzzy candidate to count as a screening HIT (delivered as a sanctions-listing record and counted in the audit verdict as "hit"). An exact substring/alias match always counts as a hit regardless of this value. Candidates scoring within 15 points below this threshold (floor 50) are still surfaced as "possible" matches in the per-name audit summary, for manual review. Default 85 is a reasonable starting point for name screening; lower it to catch more near-misses at the cost of more false positives, raise it to reduce false positives at the cost of missing more typos/variants. Ignored when matchMode is "exact".