Home / Platform
The engine

One engine. Five layers. Every document.

Docurensic isn't a single check bolted onto a viewer. It's one engine that works through every document in depth — asking five different questions and fusing the answers into a single, explainable verdict.

§01 / LAYERS
L01

Security inspection

Before we trust a file, we make sure it's safe to open — scanning for malicious payloads, hidden or obfuscated links, and embedded scripts and macros built to slip past you.

scanning…2 checks
L02

Forensic analysis

We examine the file the way a document examiner would — metadata, fonts, revision history, and the pixels themselves — to surface the fingerprints of tampering.

scanning…4 flags
L03

Reasoning

An independent AI reads the document and reasons about it — what it claims to be, whether the details hold together, and what looks off — explained in plain language.

scanning…1 note
L04

Validation

Type-aware checks confirm the document is internally consistent for what it is — required fields present, totals that reconcile, and dates and IDs that make sense.

scanning…clear
L05

External lookups

Where it helps, we corroborate against the outside world — verifying companies, domains, and identifiers — so a document isn't just internally consistent, but actually real.

scanning…1 match
Review
Altered figure detectedflagged in the forensic layer
52
Depth in five passes

Each layer asks a different question.

Most tools stop at "does it open?" Docurensic keeps going — checking that a file is safe, that it hasn't been altered, that its story holds together, that its details are valid, and that the entities on it are real.

Every finding is attributed to the layer that caught it, so you always know what was flagged and why.

SafeMalicious files, hidden links, and scripts caught first
SoundTamper fingerprints in metadata, fonts, and pixels
ConsistentFields, totals, dates, and IDs that add up
RealCompanies, domains, and identifiers verified
§02 / TOOLS
The workspace

Everything you need after the verdict.

Dig deeper, connect documents, and act — without leaving Docurensic.

Analyze

Purpose-built forensic labs.

Docurensic AI reviewThe reasoning layer, on every scan.
Compare docsWord-level diff plus font forensics.
PDF toolbox & labsRedact, merge, sign; inspect metadata, images, signatures.

Investigate

Connect the dots.

CasesDocuments that share identifiers link automatically.
Review inboxA human-decision queue with the trail kept.
URL & company analyzersVet a site or a business.

Automate

Findings into action.

Smart WorkflowsWHEN / IF / THEN on every scan.
Custom rules engineYour own detection, no code.
API & webhooksRun it from your stack.
§03 / VERDICT
The output

One score. One decision. No jargon.

Every scan lands on a 0–100 risk score and one of three verdicts — with the exact findings that drove it.

Trusted

No evidence of manipulation found — a low score, not a certificate of authenticity.

Review

Red flags worth a human look — font mismatches, revision markers, or figures that don't reconcile.

Reject

Strong forensic evidence of tampering, or an active threat like a malicious macro.

Findings are evidence for a human decision — not a verdict handed down. You stay in control of the call.

Run your first scan in the next minute.

No install, no plugin. Drop a document and read the verdict.