Skip to content
TakoAPI
Back to skills

opencr-skill

by topdu

Image & Video Generation
557Downloads reported by clawskills.sh
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Extract text from images, documents and scanned PDFs using OpenOCR - supports text detection, recognition.

Description

What This Skill Does

OpenOCR is a Python library for extracting text from images, scanned documents, and PDFs. It provides five task modes: text detection, text recognition, end-to-end OCR, VLM-based recognition for formulas and tables, and document parsing with layout analysis. Supports Chinese, English, and other languages.

Handles text, formulas, tables, and layout-aware document parsing in a single library, eliminating the need for separate specialized OCR tools.

When to Use It

  • Converting scanned PDF invoices to searchable text
  • Extracting LaTeX formulas from academic paper screenshots
  • Parsing tables from document images into structured format
  • Batch processing folders of scanned receipts
  • Converting document images to Markdown with layout preserved

Example Workflow

INPUT

User asks: Parse this scanned research paper PDF and save it as Markdown

AGENT
  1. 1

    Initialize OpenOCR with task='doc' and use_layout_detection=True

  2. 2

    Run document parsing on the PDF file

  3. 3

    Iterate over per-page results

  4. 4

    Save each page result to Markdown using doc.save_to_markdown()

OUTPUT

Markdown files saved to output directory, one per page, with text, formulas, and tables preserved in structure

Installation

clawhub install topdu/opencr-skill

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links