How do I install the Markitdown skill?

Run npx skillmds@latest add modbender/markitdown in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Markitdown safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Markitdown?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Markitdown free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Markitdown?

modbender (@modbender) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Markitdown by modbender — Agent Skill

Name: Markitdown
Rating: 4.2 (5 reviews)
Author: modbender

File contents

MarkItDown Skill

Description

MarkItDown is a Python utility developed by Microsoft (source: https://github.com/microsoft/markitdown) for converting various files and office documents to Markdown. It allows me to easily extract structured text (including tables, headers, and lists) from complex formats to better understand their content. The conversion happens locally using installed Python libraries.

Safety Note: The installation process downloads the markitdown package and its dependencies from the official Python Package Index (PyPI). Processing certain formats (like YouTube URLs) requires external network access to fetch the content. Processing local files requires access to the directory where the target files are located.

Supported Formats

Office Documents: PowerPoint (PPTX), Word (DOCX), Excel (XLSX, XLS).
PDF
Images: Text extraction (OCR) and metadata (EXIF).
Audio/Video: Speech transcription (wav, mp3, Youtube URLs) and EXIF.
Web and Text: HTML, CSV, JSON, XML.
Archives and Books: ZIP archives, EPub.

Dependencies

The skill installs the utility in a local virtual environment. Most features work out-of-the-box thanks to the markitdown[all] dependencies installed via PyPI. For specific formats (audio/video), system libraries (e.g., ffmpeg) may be required and must be installed on the host.

When to Use

When you need to read, analyze, or extract information from PDF, Word, Excel, or PowerPoint files.
When document structure is important for the response (e.g., tables or formatted lists).
If you need to extract text from audio or video files, or "read" an image.

How to Use

The virtual environment is automatically set up when the skill is installed. You must run the utility from within the skill's folder.

Conversion with Console Output (STDOUT)

Useful for small files to see the result immediately.

./.venv/bin/markitdown /path/to/file.pdf

Conversion with File Output

The best option for large documents. Save the result to a .md file and read it using the read tool.

./.venv/bin/markitdown /path/to/file.pdf -o /path/to/result.md

Example: Excel Conversion

Navigate to the skill folder (e.g., cd ~/skills/markitdown) and execute:

./.venv/bin/markitdown ~/downloads/report.xlsx -o ~/downloads/report.md

After that, you can read the resulting report.md file.

---name: markitdowndescription: MarkItDown is a Python utility from Microsoft for converting various files (PDF, Word, Excel, PPTX, Images, Audio) to Markdown. Useful for extracting structured text for LLM analysis.--- # MarkItDown Skill ## Description MarkItDown is a Python utility developed by Microsoft (source: https://github.com/microsoft/markitdown) for converting various files and office documents to Markdown. It allows me to easily extract structured text (including tables, headers, and lists) from complex formats to better understand their content. The conversion happens locally using installed Python libraries. **Safety Note**: The installation process downloads the `markitdown` package and its dependencies from the official Python Package Index (PyPI). Processing certain formats (like YouTube URLs) requires external network access to fetch the content. Processing local files requires access to the directory where the target files are located. ## Supported Formats - **Office Documents:** PowerPoint (PPTX), Word (DOCX), Excel (XLSX, XLS). - **PDF** - **Images:** Text extraction (OCR) and metadata (EXIF). - **Audio/Video:** Speech transcription (wav, mp3, Youtube URLs) and EXIF. - **Web and Text:** HTML, CSV, JSON, XML. - **Archives and Books:** ZIP archives, EPub. ## Dependencies The skill installs the utility in a local virtual environment. Most features work out-of-the-box thanks to the `markitdown[all]` dependencies installed via PyPI. For specific formats (audio/video), system libraries (e.g., `ffmpeg`) may be required and must be installed on the host. ## When to Use - When you need to read, analyze, or extract information from PDF, Word, Excel, or PowerPoint files. - When document structure is important for the response (e.g., tables or formatted lists). - If you need to extract text from audio or video files, or "read" an image. ## How to Use The virtual environment is automatically set up when the skill is installed. You must run the utility from within the skill's folder. ### Conversion with Console Output (STDOUT) Useful for small files to see the result immediately. ```bash ./.venv/bin/markitdown /path/to/file.pdf ``` ### Conversion with File Output The best option for large documents. Save the result to a `.md` file and read it using the `read` tool. ```bash ./.venv/bin/markitdown /path/to/file.pdf -o /path/to/result.md ``` ### Example: Excel Conversion Navigate to the skill folder (e.g., `cd ~/skills/markitdown`) and execute: ```bash ./.venv/bin/markitdown ~/downloads/report.xlsx -o ~/downloads/report.md ``` After that, you can read the resulting `report.md` file.

Frequently asked questions

Markitdown