Business Compliance Tools

Sanctions screening regression benchmark

Sanctions screening regression benchmark is a free browser tool that produces compare supplied baseline and candidate screening observations against exact approved synthetic cases, including paired changes and missing coverage from case reference and your policy reference. No screening API runs here.

Free. No account or payment needed.

Prepare your case and original evidence. All mechanical checks run in this browser.

Case and approved rules

Compare supplied baseline and candidate screening observations against exact approved synthetic cases, including paired changes and missing coverage.

Match exact case and version identities. Count true positives, false positives, true negatives and false negatives only for one nonerror observation. Retain missing, duplicate, error and unassigned results separately. Compare each complete baseline and candidate pair against the synthetic expectation.

Original evidence files

Up to 20 files, 10 MB each and 25 MB total. Original bytes stay in this browser until explicitly selected for separately approved account saving.

No original files selected. Source references without retained originals remain unresolved.

Evidence source references

Record exact local file IDs, source locators and customer verification. An annotation does not authenticate its contents.

Original evidence source references (0)

Task-specific review records

Approved synthetic cases (0)
Supplied screening observations (0)
Advanced exact JSON inputs

No screening API runs here. Synthetic expected labels are customer-approved fixtures. Confusion counts and alert changes do not clear real parties or establish provider semantic quality.

Sources and review boundary

  • A Framework for OFAC Compliance Commitments

    Testing, auditing, internal controls and records support sanctions compliance programs. Your supplied synthetic results do not screen real parties or prove provider execution.

  • BSA/AML Independent Testing

    Risk-based testing and workpapers support documented program review. An isolated synthetic fixture or change record does not establish monitoring effectiveness.

Earlier generic workspace data remains separate. This dedicated editor does not overwrite historical purchases or earlier storage.

What does the sanctions screening regression benchmark produce?

Compare supplied baseline and candidate screening observations against exact approved synthetic cases, including paired changes and missing coverage.

Use the task-specific inputs and exact original evidence files above to prepare and review this tool in your browser. Inspect the complete named findings and scope limits before preparing the formal pack. Save unfinished work and retain the complete recovery ZIP. Customer applicability, approvals and evidence quality require responsible review.

What do you need before you start?

  • Define the selected source population and your exact policy.
  • Keep the original files and source pointers used for every finding.
  • Resolve missing inputs and inspect unresolved findings before export.

What information do you enter?

The released tool must ask only for information needed by this bounded task.

  • Case reference, organization, subject, responsible reviewer, preparation date and selected scope.
  • Your policy reference, version, approver, approval date and exact supporting source IDs.
  • Original file IDs, names, byte lengths and SHA-256 hashes with source locators, dates, custodians and verification assertions.
  • Baseline Version
  • Candidate Version
  • Cases rows / reference
  • Cases rows / synthetic Label
  • Cases rows / scenario
  • Cases rows / expected
  • Cases rows / source IDs rows
  • Observations rows / case Reference
  • Observations rows / version
  • Observations rows / observed
  • Observations rows / provider
  • Observations rows / list Version
  • Observations rows / tested On
  • Observations rows / source IDs rows

What you can download

Every result must identify its supplied facts, assumptions and unresolved items.

Free resultIncluded downloads

Enter your information, inspect the on-screen result and keep a browser backup.

Editing and direct copies of your entered information stay free.

  • Compare supplied baseline and candidate screening observations against exact approved synthetic cases, including paired changes and missing coverage.
  • Complete findings, exact task facts, mechanical counts and unresolved evidence gaps in Word, PDF, CSV and Excel.
  • A source manifest binds the approved policy, public method sources, original file hashes and every supplied source pointer.
  • A complete recovery ZIP retains unfinished drafts, prepared results and exact originals with a hashed delivery manifest.

These supported outputs are free. No checkout is required.

Fictional review report (PDF). Editable review (DOCX). Complete fictional pack (ZIP). Requires tailoring and responsible review. Samples use the finalized renderer.

How does the sanctions screening regression benchmark work?

  1. 1
    Define the selected task

    Compare supplied baseline and candidate screening observations against exact approved synthetic cases, including paired changes and missing coverage.

  2. 2
    Record exact task facts

    Match exact case and version identities. Count true positives, false positives, true negatives and false negatives only for one nonerror observation. Retain missing, duplicate, error and unassigned results separately. Compare each complete baseline and candidate pair against the synthetic expectation.

  3. 3
    Inspect complete findings

    Review source links, exact calculations and every unresolved finding.

  4. 4
    Keep recoverable work

    Save unfinished work and exact original bytes in this browser. A complete ZIP restores the accepted workspace and original files after integrity checks. Original checkpoints and purchased archives remain separate. Restore grants no payment or execution authority. Reprepare after rejected recovery before checkout.

  5. 5
    Prepare the completed pack

    Eight core files, one member per retained original, plus any method-generated chart, synthetic data, approved draft or reporting artifact. The real component calculates the exact prepared count. Downloads are free.

Common questions

What information does the sanctions screening regression benchmark need?

Enter case reference, organization, subject, responsible reviewer, preparation date and selected scope; your policy reference, version, approver, approval date and exact supporting source IDs; original file IDs, names, byte lengths and SHA-256 hashes with source locators, dates, custodians and verification assertions; baseline Version; candidate Version. Define the selected source population and your exact policy.

What does the sanctions screening regression benchmark produce?

Compare supplied baseline and candidate screening observations against exact approved synthetic cases, including paired changes and missing coverage. Complete findings, exact task facts, mechanical counts and unresolved evidence gaps in Word, PDF, CSV and Excel. A source manifest binds the approved policy, public method sources, original file hashes and every supplied source pointer. A complete recovery ZIP retains unfinished drafts, prepared results and exact originals with a hashed delivery manifest. Review unresolved findings and limitations before use.

Who makes the final decision?

A responsible person reviews the result. No screening API runs here. Synthetic expected labels are your fixtures. Confusion counts and alert changes do not clear real parties or establish provider semantic quality.

Is this tool free?

Yes. The on-screen result and listed downloads are free. No checkout is required. Keep local backups and review the stated method and limits.

Can I use the sanctions screening regression benchmark for this task?

Compare screening results against approved synthetic cases and retain missing coverage. No screening API runs here. Synthetic expected labels are your fixtures. Confusion counts and alert changes do not clear real parties or establish provider semantic quality.

What still needs review?

No screening API runs here. Synthetic expected labels are your fixtures. Confusion counts and alert changes do not clear real parties or establish provider semantic quality.