Documents in.
Structured truth out.
Parse, extract, audit and translate any document — PDF, Word, scans, photos. Production-grade accuracy, 100-page free trial, no credit card.
Free tier includes 100 pages · cancel anytime
- contract-v3.docx
- invoice-2026.pdf
- scan-004.jpg
- Parse
- Extract
- Audit
- Translate
"summary": { "high": 3, "medium": 1 }, "findings": [ { "rule": "unlimited-liability", "page": 4, "severity": "high" }, ... ]
One pipeline, four superpowers
Every document runs through the same deterministic pipeline — parse once, then extract, audit and translate on demand.
Parse
PDF, Word, scans & photos to clean Markdown with layout, tables and reading order preserved.
Extract
Define a schema, get clean JSON or Excel. Field-level provenance links every value to its source page.
Audit
12 risk rules for CN & EN contracts — unlimited liability, auto-renewal, unilateral terms — each finding cited to its clause.
Translate
EN ↔ 中文 (Simplified & Traditional) with layout intact. Char-level OpenCC for perfect 简繁 conversion.
Built for teams that live in documents
Finance & Ops
Invoices, POs, statements — straight into your ERP. Line items with page-level provenance.
Legal & Compliance
Contract audits in seconds. Flag unlimited liability, auto-renewal and unilateral clauses before you sign.
Global Teams
Cross-border documents translated with layout intact — 简繁 conversion that never corrupts terms.
One REST call.
Zero glue code.
Idempotent jobs, signed webhooks, SDKs for Python & Node. Pay per page, never per seat.
# pip install docxstack from docxstack import Client client = Client() # uses session cookie job = client.parse("contract.pdf") report = client.audit(job) for f in report.findings: print( f.severity, f.title, f"page {f.page}") # high Unlimited / uncapped liability page 4 # high Auto-renewal with opt-out burden page 6
Start reading documents like data.
100 free pages. No credit card. Your first audited contract is three clicks away.