ocr

Extract text and structured fields from images and scanned documents.

4|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/JigsawStack/interfaze-skills --skill ocr-jigsawstack
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ocr
Source: https://github.com/JigsawStack/interfaze-skills/tree/main/skills/ocr
Command: npx skills add https://github.com/JigsawStack/interfaze-skills --skill ocr-jigsawstack

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

OCR removes the manual effort of reading and retyping information from screenshots, scans, receipts, invoices, forms, and complex documents by turning visual content into usable text or structured fields.

Core Features & Use Cases

  • Visual OCR for complex layouts: Extracts text from images and scanned documents while respecting layout where needed.
  • Structured field extraction: Produces schema-constrained outputs (e.g., Zod/Pydantic) for fields like names, addresses, totals, line items, or IDs.
  • Plain or raw OCR modes: Supports plain text extraction and a raw mode that returns detailed OCR artifacts such as bounding boxes and confidence signals.

Quick Start

Ask the AI to extract the text and key fields from your uploaded receipt image so you get the readable output immediately.

Frequently Asked Questions about ocr

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract text from a scanned invoice image?

OCR extracts text from scanned invoice images by digitizing visual content into usable text. It captures readable output and structured fields like names, totals, and line items directly from the image without manual retyping.

Can I extract specific fields from receipts instead of plain text?

Yes, receipt parsing supports schema-constrained outputs alongside plain text mode. You can define schemas using Zod or Pydantic to extract specific structured fields like vendor names, addresses, totals, and line items from the visual input.

Does this OCR tool work with PDF files and screenshots?

Yes, the OCR tool processes both PDF file inputs and screenshots. It applies layout-aware information extraction to scanned documents and images, ensuring readable output from complex visual inputs while respecting the original document layout.

What is raw mode for document digitization and when do I need it?

Raw mode in document digitization returns detailed OCR artifacts such as bounding boxes and confidence signals alongside the text. You need it when building applications that require precise text location data or extraction accuracy verification.

How do I get structured data from an ID card image?

You get structured data from an ID card image by applying schema-based OCR extraction. The tool processes the visual input and returns constrained output fields, converting the physical card information into usable digital data formats.