markitdown

Converts PDFs and Office files to Markdown for agents

★ 192k · updated yesterday · Official, published by Microsoft

View microsoft/markitdown on GitHub

Skills can run commands on your computer. Review a skill before installing.

Listed from public GitHub data. Not affiliated with or endorsed by the authors unless marked Official. Names and logos belong to their owners and only identify the repositories. Sources and removal requests

What it does

A Python tool and library from Microsoft that converts files such as PDF, Word, PowerPoint, Excel, HTML and images into Markdown, a format language models read well. It also comes with an MCP server, so an agent can convert files itself.

Read the README on GitHub

Who it’s for

Anyone who feeds documents to an AI agent or a RAG pipeline and wants clean text instead of raw files.

Install

Install it with:

pip install 'markitdown[all]'
markitdown path/to/file.pdf > file.md

Run it with --help to see its commands.

Activity

Last commit
yesterday
Open issues and pull requests
757
Forks
14,323

From GitHub. Stars and commits are refreshed daily; contributors, languages and releases weekly.

At a glance

Licence: MIT. The licence is the authors’, as GitHub reports it. Check the repository before you rely on it.

Type
CLI tool
Languages
Python
Maintainer
microsoft
Stars on GitHub
★ 192k
Badges
Top, Official (Microsoft)

Is this your repository? Choose “Request removal” on this page, or email legal@antspace.ai: we act on removal requests within 48 hours.

Similar skills

More in Research & documents · All AI skills