Free document toolsPOST /v1/adverse-media

Screen candidate documents before you hire

Supported Formats: Identity Documents (100-point), Payslips & Income Evidence, Bank Statements, Candidate Name & DOB
Core Verification Mechanisms: Sanctions Screening · Corroboration-Gated Media · 100-Point Identity Tally · Pack Checklist
Need copy-paste code with sandbox testing?Open Developer Playbook (cURL / Python / MCP) →

The problem

Hiring for a regulated or high-trust role means screening against sanctions lists and adverse news — then proving the person is who the paperwork says.

Manual background checks take days, and a naive name search buries a real hit under a hundred people who share the name.

How Stipple helps

Screen the candidate name with entity disambiguation: sanctions list matching (PEP is not screened, and the result says so), and same-name articles pushed into an excluded bucket rather than counted as hits.

Run their identity documents through the AFP 100-point scheme in the same workflow — a deterministic points tally showing which documents counted and how far short a pack falls.

bash
curl -X POST https://www.stipple.sh/v1/adverse-media \
  -H "Content-Type: application/json" \
  -d '{"name": "Jordan Sample", "country": "AU"}'
# → { "risk_flag": "clear", "sanctions": { "matches": 0 }, ... }

Screen candidates with auditable evidence in seconds without manual database trawling.

Questions

Can this tell me whether a candidate has the right to work?
No. There is no right-to-work determination here, and no immigration or visa database behind it. What you get is an identity result — the AFP 100-point tally showing which documents were recognised and what they were worth — plus a name screen for sanctions and adverse media. Establishing work rights is a separate check against the relevant government service, and this does not replace it.
Does it verify a degree or a professional certification?
It does not contact an issuing institution, so it cannot confirm a qualification was awarded. A certificate is inspected as a document — was it tampered with, does it hold up structurally — and that is a different question from whether the university has a record of it. Treat a clean document result as one input, not as verification of the credential.
How do you avoid flagging every candidate who shares a name with someone notorious?
By refusing to count a name match as a hit. Sanctions candidates are corroboration-gated: a name-only match is reported as possible, never confirmed, because one common name belongs to several different people. Adverse-media results are entity-disambiguated, and same-name articles are surfaced separately under excluded rather than folded into the count. You can sharpen it further by passing date of birth, country, employer, or role.
Is a clear result a clean background check?
No, and treating it as one is the mistake this is designed to prevent. Clear means these sources found nothing, which is a statement about coverage. It is not a finding of good character, and it is not a substitute for a formal police check where the role requires one.
What does corroboration-gated adverse media mean?
A media hit is only surfaced when it corroborates against the submitted identity pack — name and date-of-birth alignment — which removes most of the same-name noise that makes raw media screening unusable. Excluded same-name hits are still shown, so you can see what was set aside and why.

Sources

  1. 01Developer playbook: the 100-point identity check
  2. 02Developer playbook: checking a document pack for completeness