Document OCR & Extract

Turn scanned PDFs into clean structured fields.

Open source · Dify template · Apache-2.0

Overview

Reads scanned or photographed documents and returns structured JSON — fields, tables and line items — ready to push into your ERP, spreadsheet or database. Handles invoices, forms and purchase orders.

What it does

OCR for scans and phone photos
Configurable field schema per document type
Table and line-item extraction
Confidence score per field for human review

Works with

Google DriveWebhook / REST API

Deploy this agent

We deploy and configure this agent in your environment, connect it to your tools, and tune it to your policies.

Book a demo Request access

How deployment works

1. Connect your tools via secure OAuth — no data leaves your tenant.
2. We tune the agent to your policies and templates.
3. Go live with a human in the loop, then automate.