API Documentation

Screen names against the OFAC SDN, EU consolidated, UK OFSI and UN sanctions lists. The API is distributed through RapidAPI — subscribe there to get your key, then send it with each request.

Authentication

RapidAPI injects your credentials automatically. From your code you call the RapidAPI endpoint with your X-RapidAPI-Key; RapidAPI forwards the request to the backend over a private secret. You never handle the backend secret yourself.

curl --request GET \
  --url 'https://<host>.p.rapidapi.com/v1/screen?name=Vladimir%20Putin' \
  --header 'X-RapidAPI-Key: YOUR_KEY' \
  --header 'X-RapidAPI-Host: <host>.p.rapidapi.com'

GET /v1/screen

Screen a single name.

Param	Type	Default	Notes
name	string	—	required, 1–200 chars
threshold	int	82	min score 50–100
limit	int	10	max entities 1–50

{
  "query": "Vladimir Putin",
  "query_normalized": "vladimir putin",
  "is_match": true,
  "match_count": 2,
  "matches": [{
    "score": 100.0,
    "confidence": "very_high",
    "matched_name": "Vladimir PUTIN",
    "matched_name_type": "aka",
    "exact": true,
    "evidence": {
      "token_sort_ratio": 100.0, "token_set_ratio": 100.0,
      "query_coverage": 1.0, "name_coverage": 1.0, "shared_tokens": 2
    },
    "entity": {
      "primary_name": "Vladimir Vladimirovich PUTIN",
      "type": "individual", "source": "OFAC_SDN",
      "programs": ["RUSSIA-EO14024"], "dobs": ["07 Oct 1952"],
      "nationalities": ["Russia"]
    }
  }],
  "elapsed_ms": 41.2
}

POST /v1/screen/batch

Screen up to 100 names in one request — ideal for customer onboarding or periodic re-screening of an existing book.

POST /v1/screen/batch
{ "names": ["Kim Jong Un", "Maria Garcia"], "threshold": 82, "limit": 5 }

// response
{ "count": 2, "matched": 1, "clean": 1, "elapsed_ms": 88.0, "results": [ ... ] }

GET /v1/lists

Coverage and freshness metadata — which lists are loaded, entity counts and when they were last refreshed. Useful for your own audit logs.

Match scoring

Scores are 0–100. A two-stage engine retrieves candidates by token-set similarity, then re-scores each with a coverage-weighted composite so that a short list entry cannot spuriously match a long query. Confidence bands: very_high ≥97, high ≥90, medium ≥82. Tune threshold to trade recall for precision.

Get your API key