What file formats can MinerU OCR handle?

It supports a wide range of formats including PDF, DOC, DOCX, PPT, PPTX, and image formats like PNG, JPG, and JPEG.

Where are the converted files stored?

Converted Markdown files are saved in the same directory as the source file, and a backup archive is maintained within the skill's archive directory.

Can it handle mathematical equations?

Yes, by enabling the MINERU_ENABLE_FORMULA option in the configuration, the skill can recognize and convert math formulas.

How long does the API Token last?

The MinerU API Token typically has a validity period of 14 days, after which you will need to generate a new token and update your configuration.

Does this skill require an external API key?

Yes, you must obtain a MinerU API Token from the official MinerU website (mineru.net) and configure it in your environment settings to use this skill.

MinerU Document OCR & Markdown

Name: MinerU Document OCR & Markdown
Author: cat-xierluo

bycat-xierluo

•

生産性とワークフロー

Converts PDFs, images, and office documents into structured Markdown using high-accuracy OCR and layout recognition.

This skill integrates the MinerU API into Claude Code to seamlessly transform complex documents—including PDFs, Word files, and PPTs—into clean, editable Markdown format. It excels at preserving document structure by accurately identifying and converting tables, mathematical formulas, and scanned text through advanced OCR. It is particularly valuable for legal professionals, researchers, and developers who need to digitize paper-based or layout-heavy documents for further processing, analysis, or documentation purposes.

主な機能

01Customizable polling and language settings

0256 GitHub stars

03High-accuracy OCR for scanned PDFs and images

04Mathematical formula recognition and conversion

05Preservation of complex table structures and layout

06Support for PDF, DOCX, PPTX, and JPEG formats

ユースケース

01Converting academic papers with complex tables into Markdown

02Digitizing scanned legal documents for searchable archives

03Extracting content from legacy office files for LLM processing

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cat-xierluo/legal-skills mineru-ocr

For use in Claude.ai and ChatGPT

Download Skill