The extraction layer for products that read documents. Your users upload a file — a trade license, an invoice, a passport — and your app receives structured data in seconds. One API. Any field. No pipeline to maintain.
Crooked phone shot, blurry scan, 12-page PDF — same API call. Docex handles orientation, OCR, retries, and schema validation so your product doesn't have to.
await docex.extract({ file: "./uploaded-license.heic", prompt: "company name, number, expiry", }); // → 200 OK · 2.4s · $0.05 { "legal_name": "ACME LOGISTICS L.L.C", "license_no": "1019388", "expires_on": "2026-03-13" }
Describe what you need. Docex reads the rest.
Line items, totals, tax, vendor, due dates
Legal name, number, expiry, activity
Name, number, nationality, dates
Merchant, amount, date, category
Parties, terms, clauses, signatures
Balance, transactions, account details
Billed at 2× upstream cost. Drop in five dollars, extract about a hundred pages, and keep going. No annual commit, no sales call, no pricing spreadsheet. Infrastructure pricing for product teams.
Paste this into your coding agent. It wires Docex into your product, scaffolds the endpoint, and runs a smoke test. You approve one link. Five minutes to production-ready extraction.
Wire Docex into this project as the document extraction layer for [describe the use case — e.g., "reading trade licenses during user onboarding"]. Take me through the GitHub approval and the $5 wallet top-up, store the API key in my env, scaffold a server-side extraction endpoint for my stack, and run a smoke test to confirm the integration works end-to-end.