StippleDocs

Examples by use case

The same verification suite, mapped to the workflows people actually run it in. Each example is a single HTTP call; follow the walkthrough for the problem, the full response, and how to wire it in.

Want to see real output first? The sample KYC onboarding report shows a complete, explainable result — points tally, per-document scorecard, and the exact gap on an incomplete pack.

Business workflows

KYC & customer onboarding

Run the identity gate over a customer’s document pack. The Australian 100-point check returns whether identity is established and exactly what’s missing if not.

bash
curl -X POST "https://www.stipple.sh/v1/identity-check?scheme=afp_100_point" \
  -F "files=@passport.pdf" -F "files=@licence.pdf"
# → { "established": true, "points": 110, "target": 100,
#     "documents": [ { "classified_type": "australian_passport", ... } ] }

POST /v1/identity-check · Full walkthrough →

Lending & income verification

Forensically check a payslip or bank statement for tampering and arithmetic before a credit decision.

bash
curl -X POST https://www.stipple.sh/v1/warrants -F "file=@payslip.pdf"
# → { "risk_band": "high", "recommended_action": "review_before_action",
#     "signals": [ { "id": "domain.payslip.reconciliation", ... } ] }

POST /v1/warrants · Full walkthrough →

Insurance claims

Check a repair invoice or quote for tamper signals and arithmetic reconciliation before a payment is authorised.

bash
curl -X POST https://www.stipple.sh/v1/warrants -F "file=@repair-invoice.pdf"
# → { "risk_band": "high", "recommended_action": "review_before_action", ... }

POST /v1/warrants · Full walkthrough →

Rental applications

Check a rental pack for completeness against the rental-application checklist, and verify the income documents in the same pass.

bash
curl -X POST "https://www.stipple.sh/v1/check-pack?scheme=rental_application" \
  -F "files=@payslip.pdf" -F "files=@bank-statement.pdf" -F "files=@licence.png"
# → { "complete": false, "missing": ["proof_of_address"],
#     "documents": [ { "classified_type": "payslip", ... } ] }

POST /v1/check-pack · Full walkthrough →

HR & credential screening

Screen a candidate’s name for adverse media and sanctions/PEP exposure — corroboration-gated, with same-name false matches filtered out.

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

POST /v1/adverse-media · Full walkthrough →

Content & agents

Fact-check an AI report

Resolve and verify every citation in an AI-generated report and recompute its internal arithmetic, before it reaches a client.

bash
curl -X POST https://www.stipple.sh/v1/verify-references \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/report.pdf"}'
# → { "summary": { "headline": "60/63 citations resolve; 3 dead links" },
#     "permalink": "https://www.stipple.sh/fc/..." }

POST /v1/verify-references · Full walkthrough →

Detect AI-written text

Estimate the probability that prose was AI-written, with the specific tells — abstaining on non-prose where detection is unreliable.

bash
curl -X POST https://www.stipple.sh/v1/detect-ai-text \
  -H "Content-Type: application/json" \
  -d '{"text": "The submitted essay text..."}'
# → { "probability": 0.92, "lean": "ai", "tells": [...], "applicable": true }

POST /v1/detect-ai-text · Full walkthrough →

Verification inside an agent

Give an MCP-capable agent the full verification suite over one endpoint — it calls the tools mid-task.

bash
claude mcp add --transport http openwarrant https://www.stipple.sh/mcp

# then, in conversation:
# "Verify this onboarding pack: <files>"

MCP (Streamable HTTP) · Full walkthrough →

Next steps