Use case

Verify claim documents before you pay

The problem

A claim arrives with an invoice, a couple of quotes, and some photos. The documents look ordinary — and anyone can now generate a clean repair invoice from a business that does not exist.

Assessors work under volume and time pressure, so the consistent-looking fakes — plausible totals, tidy layout, matching letterhead — are exactly the ones manual review waves through.

How Stipple helps

Stipple runs forensic inspection on each document: file-structure and metadata analysis, font and layout integrity, image-tamper signals, and arithmetic reconciliation — do the line items actually sum to the total, with the right tax?

You get a risk band, the specific findings that fired, and an audit-ready report — a signal that routes the suspicious claim to a human with the evidence already highlighted, not a black-box verdict.

The call

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

Screen at first notice of loss: clean claims flow through, and the ones with tamper or arithmetic signals reach an assessor before the payment is authorised.