Markdownify MCP Server

What is Markdownify MCP Server about?

Markdownify MCP Server provides a suite of conversion tools that transform a wide range of file formats and web content into Markdown. It runs as a Model Context Protocol (MCP) server, exposing endpoints that can be called from desktop applications or other MCP‑compatible clients.

How to use Markdownify MCP Server?

Clone the repository

git clone https://github.com/zcaceres/markdownify-mcp.git
cd markdownify-mcp

Install dependencies (Node.js, pnpm, plus Python dependencies via uv)
```
pnpm install
```
Build the TypeScript source
```
pnpm run build
```
Start the server
```
pnpm start
```

Integrate with a desktop app by configuring the MCP server entry, e.g.:

{
  "mcpServers": {
    "markdownify": {
      "command": "node",
      "args": ["/absolute/path/to/dist/index.js"],
      "env": { "UV_PATH": "/path/to/uv" }
    }
  }
}

The server will then respond to tool calls such as pdf-to-markdown, image-to-markdown, youtube-to-markdown, etc.

Key features of Markdownify MCP Server

Multi‑format conversion: PDF, images, audio (with transcription), DOCX, XLSX, PPTX.
Web content extraction: YouTube video transcripts, Bing search results, arbitrary web pages.
File retrieval: Serve existing Markdown files; optional MD_SHARE_DIR env var restricts the accessible directory.
Extensible toolset: All tools are defined in src/tools.ts; developers can add new converters.
Cross‑language support: Calls can be made from any MCP‑compatible client (desktop apps, web services, etc.).

Use cases of Markdownify MCP Server

Documentation pipelines – automatically convert design docs, reports, and presentations into Markdown for version‑controlled wikis.
Research assistants – pull YouTube lecture transcripts or Bing search snippets directly into a Markdown note.
Content aggregation – gather data from multiple sources (PDFs, spreadsheets, web pages) into a single, readable Markdown report.
AI prompt engineering – supply clean Markdown inputs to LLMs for better context comprehension.

FAQ from the Markdownify MCP Server

Do I need Python installed? The project uses uv to manage Python dependencies; pnpm install will download and set up uv automatically.
Can I run the server on Windows? Yes, though the repository currently lacks Windows‑specific CI; contributors are welcome to add and test Windows support.
How do I restrict which Markdown files can be served? Set the MD_SHARE_DIR environment variable before starting the server to limit file access to a specific directory.
What if I want to add a new conversion tool? Add the implementation in src/tools.ts and expose it through the server’s tool registry.
Is the server open‑source? The code is released under the MIT License; feel free to fork, modify, and contribute.

Help! I need someone with a Windows computer to help me add support for Markdownify-MCP on Windows. PRs exist but I cannot test them. Post here if interested.

Markdownify is a Model Context Protocol (MCP) server that converts various file types and web content to Markdown format. It provides a set of tools to transform PDFs, images, audio files, web pages, and more into easily readable and shareable Markdown text.

Features

Convert multiple file types to Markdown:
- PDF
- Images
- Audio (with transcription)
- DOCX
- XLSX
- PPTX
Convert web content to Markdown:
- YouTube video transcripts
- Bing search results
- General web pages
Retrieve existing Markdown files

Getting Started

Clone this repository
Install dependencies:
```
pnpm install
```

Note: this will also install uv and related Python depdencies.

Build the project:
```
pnpm run build
```
Start the server:
```
pnpm start
```

Development

Use pnpm run dev to start the TypeScript compiler in watch mode
Modify src/server.ts to customize server behavior
Add or modify tools in src/tools.ts

Usage with Desktop App

To integrate this server with a desktop app, add the following to your app's server configuration:

{
  "mcpServers": {
    "markdownify": {
      "command": "node",
      "args": [
        "{ABSOLUTE PATH TO FILE HERE}/dist/index.js"
      ],
      "env": {
        // By default, the server will use the default install location of `uv`
        "UV_PATH": "/path/to/uv"
      }
    }
  }
}

Available Tools

youtube-to-markdown: Convert YouTube videos to Markdown
pdf-to-markdown: Convert PDF files to Markdown
bing-search-to-markdown: Convert Bing search results to Markdown
webpage-to-markdown: Convert web pages to Markdown
image-to-markdown: Convert images to Markdown with metadata
audio-to-markdown: Convert audio files to Markdown with transcription
docx-to-markdown: Convert DOCX files to Markdown
xlsx-to-markdown: Convert XLSX files to Markdown
pptx-to-markdown: Convert PPTX files to Markdown
get-markdown-file: Retrieve an existing Markdown file. File extension must end with: *.md, *.markdown.

OPTIONAL: set MD_SHARE_DIR env var to restrict the directory from which files can be retrieved, e.g. MD_SHARE_DIR=[SOME_PATH] pnpm run start

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

{ "mcpServers": { "markdownify": { "command": "node", "args": [ "/absolute/path/to/dist/index.js" ], "env": { "UV_PATH": "/path/to/uv" } } } }

Markdownify MCP Server