OpenSanctions
Open data and screening infrastructure
OpenSanctions aggregates entity and watchlist data and exposes matching through a hosted API or the self-hosted yente service. It is a data and matching foundation, not a complete compliance department workflow. Commercial use requires the appropriate data license even when the software is self-hosted.
Our assessment
A transparent screening foundation. Budget for the operating system around it.
Best fit and limitations
A strong infrastructure candidate for teams able to build or already operating screening review, case management, and audit controls. Transparency and deployment control are the main evaluation advantages.
- Open-source software does not make commercial data use free.
- No complete native case management, report filing, or AML transaction monitoring is established here.
- Do not assume broad commercial adverse-media coverage from the presence of crime-related public records.
Tools 5
OpenSanctions datasetsRisk data+
- Select individual sources or combined collections
- Inspect entities and source-linked properties
/match APIEntity screening+
- Query by example with names and additional identifiers
- Return ranked candidate matches for review
yenteSelf-hosted matching+
- Run the matching API on your own infrastructure
- Use selected OpenSanctions and internal data
Matching configurationPrecision and recall control+
- Choose algorithms and feature weights
- Separate candidate retrieval from match scoring
/search APIAnalyst search+
- Full-text exploration of entities
- The provider explicitly recommends /match for screening
AI capabilities
Transparent matching, not an AI investigator
The recommended logic-v2 matcher uses deterministic rules, culturally aware name matching, and identifier checks. It returns candidates and similarity scores; it is not an AML investigator or a predictive fraud model.
What to validate
Version datasets and algorithm settings. Validate thresholds and keep analyst decisions independent from raw similarity scores.
Implementation
Integration checklist
- Use /match for screening and /search for human exploration.
- Implement rescreening, evidence snapshots, decisions, audit logs, and update monitoring.
- For self-hosting, operate data refresh, indexing, backup, monitoring, and capacity.
Commercial scope
Commercial data license required. Hosted API and self-hosted operation have different costs; verify current pricing and data rights directly.
Questions for the demo
- Which datasets and refresh schedules satisfy our specific screening scope?
- What commercial rights cover hosted, self-hosted, and derived use?
- How will we prove a past decision against the exact data and matching version?
Engineering
- Inputs
- Query-by-example person or organization attributes · Entity properties, selected algorithm, feature weights, and threshold
- Outputs
- Ranked candidate entities from the selected data collection · Ranked candidate entities and matching scores
- Webhooks
- Not verified in public documentation.
- Decision timing
- No measured latency or contracted response time verified.
- Deployment
- Hosted API · On-premises
- Dependencies
- Commercial data license for business use · Explicit dataset, algorithm, and threshold policy · Pin the algorithm when changes in matching behavior require validation · Business use requires a commercial license unless the provider grants an applicable exemption
Verified details 10
- Inputs
Query-by-example person or organization attributes
Source Checked 2026-09-18 - Outputs
Ranked candidate entities from the selected data collection
Source Checked 2026-09-18 - Deployment
Hosted API
Source Checked 2026-09-18 - Deployment
On-premises
Source Checked 2026-09-18 - Dependency
Commercial data license for business use
Source Checked 2026-09-18 - Inputs
Entity properties, selected algorithm, feature weights, and threshold
Source Checked 2026-09-18 - Outputs
Ranked candidate entities and matching scores
Source Checked 2026-09-18 - Dependency
Explicit dataset, algorithm, and threshold policy
Source Checked 2026-09-18 - Dependency
Pin the algorithm when changes in matching behavior require validation
Source Checked 2026-09-18 - Dependency
Business use requires a commercial license unless the provider grants an applicable exemption
Source Checked 2026-09-18
Vendor example
Published vendor example. Confirm the current version and required credentials.
{
"schema": "Person",
"properties": {
"firstName": [
"Arkadii"
],
"fatherName": [
"Romanovich"
],
"lastName": [
"Rotenberg",
"Ротенберг"
],
"birthDate": [
"1951"
],
"nationality": [
"Russia"
]
}
}Integration limits
- Published vendor payload example, linked to its source. It is not a complete integration or a tested production request.
- Empty country, deployment, or provider lists mean not verified in this review. A documented country refers to the specific product noted in its source, not universal platform coverage.
- The matcher supplies evidence for a screening process. The client must implement review, rescreening, audit retention, and operational decisions.
- The vendor example is the published entity object, not a complete HTTP request. The source's trailing comma was removed to make valid JSON; field values are unchanged.
- Pin an algorithm for reproducible behavior. Selecting best can change the model and scores as the publisher updates its default.
- The recommended logic-v2 matcher and older regression algorithms have different behavior. Do not interpret a score as a calibrated sanction probability.
- Open-source hosting software and licensed data are separate rights.
Sources 5
Source review dates are shown above. Product claims come from public sources. Fit, limits, and evaluation questions are our analysis. This is not a hands-on performance test. Methodology · Changelog