Home/Guides/Document Verification API
For Developers
POST a document. Get a verdict with named findings. No sales call, no template setup.
One REST endpoint for document fraud detection: upload a PDF, scan, or photo, get a structured JSON verdict with a calibrated risk score and specific findings tied to document regions. Covers 100+ document types, stores nothing, and is billed per check.
- No document storage
- Digital, scanned or photographed
- ~1 minute per verdict
Last updated:
The problem
Why document verification belongs in the API layer
Most document fraud tooling is built for enterprise procurement: demo calls, annual contracts, per-seat dashboards. Meanwhile the actual work - deciding whether a submitted PDF is genuine - happens inside your onboarding, lending, or claims pipeline, where what you need is an endpoint, not a portal.
Building verification in-house doesn't scale either. One or two forensic checks are a weekend project; maintaining 200+ checks against fraudsters who update their tooling every few months is a permanent team.
An API-first verification layer means fraud detection sits inline in your existing flow: documents are checked at upload, verdicts route the application, and suspicious files never reach downstream systems as trusted data.
What we detect
Forensic signals that expose fraud
Single REST endpoint
POST the file to /api/v1/documents/ with your API key. No SDKs required, no callback gymnastics for the simple path - an async job that returns a verdict within seconds, with webhooks available for batch workflows.
Structured JSON verdicts
Every response includes a verdict (authentic, suspicious or tampered), a calibrated 0-100 risk score, and itemised findings tied to regions of the document - ready to route programmatically.
200+ forensic checks per call
Font metrics, running-balance arithmetic, metadata fingerprints, text-layer vs raster comparison, ELA, AI-generation detection, and more - all running in parallel on every submission.
100+ document types
Bank statements, payslips, passports, driver's licences, utility bills, academic credentials, invoices, trade licences - across most major issuing countries, with no per-template configuration.
Zero document storage
Documents are analysed in memory and discarded once the verdict is returned. Nothing is retained - which keeps GDPR and data-minimisation reviews short.
Managed AI inference
The adjudication layer runs on TamperCheck-managed infrastructure with audited models. No provider account to set up, no keys to rotate, and nothing extra for your security review to cover.
The solution
Self-serve, API-first, documented. No sales call.
Create an account, generate an API key, make your first call with matched trial credits. The verdict format is stable, documented, and built to be routed on.
How it works
- One endpoint - POST /api/v1/documents/ with the file, get JSON back
- Verdict in about a minute - fast enough to sit inline in onboarding
- Webhooks for asynchronous and batch pipelines
- per-document pricing - usage-based, no subscriptions or minimums
- Trial credits matched free on signup - test with your own documents first
- Full API docs at tampercheck.ai/docs - no sales call required
Why us
What you won't get elsewhere
Findings, not a score
We name the field that changed and where it sits on the page.
Others hand you a number to justify.
One endpoint, every type
100+ document types, 190+ countries, no per-type setup.
Elsewhere, each new form is a project.
Catches what was never real
A dedicated pass for AI-generated and deepfake documents.
Template matching only catches edits.
Nothing is retained
Analysed in memory, then discarded. Never trained on.
Most services keep documents for QA.
Live in minutes
Published pricing, no minimum. Test it before you talk to us.
Platforms want a contract first.
Sits alongside your stack
Keep Onfido or Jumio for face match. We add the forensics.
Not a rip-and-replace.
FAQ
Common questions
What is a document verification API?
An API that accepts a document (PDF, scan, or photo) over HTTP and returns a machine-readable judgment on whether it's genuine - covering tampering, forgery, and AI generation - so fraud checks run inline in an automated workflow instead of a manual review queue.
How fast is the API?
A full analysis returns in about a minute per document. Checks run in parallel, so complex documents don't take proportionally longer.
What does the response look like?
A JSON object with a verdict (authentic, suspicious or tampered), a 0-100 risk score calibrated per document type, and an array of findings - each with a description and the document region it applies to.
Is there a free tier or trial?
New accounts double their first top-up: we credit a matching amount free, so you start with double the balance at the standard per-document rate. A card is required to activate, and there are no subscriptions or monthly minimums after that.
Do you store the documents we send?
No. Files are analysed in memory, the verdict is returned, and the bytes are discarded. Zero storage by design - so integrating the API doesn't add a new data-retention surface to your compliance review.
Can we use our own AI provider keys?
No - AI inference is managed by TamperCheck on audited models. No provider accounts to connect, no keys to rotate. Per-document pricing already covers inference, so there's no separate provider bill.
Or skip the call entirely
We'll match your first top-up in credits, so you can run your own documents through it now.