opendatalabopendatalabOfficialยท1 Agent Skills Included

MinerU-Ecosystem

Convert PDFs, Office files, and web pages to Markdown

Converts PDFs, scanned documents, images, Word, Excel, PowerPoint, and web pages into clean Markdown, HTML, LaTeX, or DOCX. Eliminates garbled tables, broken formulas, and unreadable OCR output with a dual VLM and OCR engine supporting 109 languages. Offers a free no-login flash mode for quick files and a precision mode with batch processing for large document pipelines.
npx skills add opendatalab/MinerU-Ecosystem --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install MinerU-Ecosystem?โ–ผ

Run `npx skills add opendatalab/MinerU-Ecosystem --all -g -y` in your terminal to install all skills in this suite globally.

How to convert a PDF to Markdown?โ–ผ

Run `mineru-open-api flash-extract report.pdf` for instant conversion with no login, or use `mineru-open-api extract` for higher accuracy and extra formats like DOCX and LaTeX.

Can MinerU handle scanned documents and tables?โ–ผ

Yes. Its VLM plus OCR dual engine recognizes scanned pages, handwriting, complex tables, and math formulas across 109 languages.

Does MinerU work with Claude and Cursor?โ–ผ

Yes. It ships an MCP server and a standard SKILL.md, so it runs natively in Claude Desktop, Cursor, Windsurf, and other compatible clients.

Do I need an API key to use MinerU?โ–ผ

No. Flash extract mode works free with no token for files up to 10 MB or 20 pages; a free token unlocks batch processing and larger files.

Related Repositories in Data & Analytics

View All in Data & Analyticsโ†’