markdownify-mcp Lets AI Agents Read PDFs, Audio, and Videos on Demand
markdownify-mcp is a free, open-source MCP server that lets AI agents convert PDFs, images, audio, Office docs, and web pages into Markdown on demand
Takeaways
- What it is: markdownify-mcp is a free, MIT-licensed MCP server that converts PDFs, images, audio, Office docs, and web pages to Markdown.
- 10 tools included: Covers PDF, image (OCR), audio (transcription), DOCX, XLSX, PPTX, YouTube, Bing search, web pages, and full git repos.
- Built on MarkItDown: Uses Microsoft's open-source MarkItDown Python library under the hood, wrapped in a TypeScript MCP layer.
- Key limitation: Scanned (image-only) PDFs return near-empty output; the Docker slim image also skips audio and image OCR.
- Setup: Clone, run
bun install(sets up Python venv automatically), build, and point your MCP client config at the built index.js. - Access controls: Set
MD_ALLOWED_PATHSto restrict which directories the server can read from, important for sandboxed or shared deployments.
markdownify-mcp is a small but surprisingly powerful open-source tool that solves one of the most persistent friction points in AI-assisted workflows: your data is never in the right format. PDFs, slide decks, audio recordings, spreadsheets, YouTube videos , none of them are natively readable by an LLM without some preprocessing step. This MCP server makes that step disappear, letting your AI agent convert almost any content type to Markdown mid-conversation, on demand.
The format problem no one talks about
If you're building an AI-assisted workflow, you eventually hit the same wall: your inputs aren't Markdown. That's the core problem markdownify-mcp addresses , it's a Model Context Protocol (MCP) server that exposes a set of "convert this thing into Markdown" tools. The reason Markdown matters so much here is well-established: LLMs appear to have been trained on vast amounts of Markdown-formatted text and understand it well. As a side benefit, Markdown conventions are also highly token-efficient.
MCP itself , short for Model Context Protocol , is an open protocol introduced by Anthropic in late 2024 to standardize how AI applications provide context to LLMs. In essence, MCP defines a common client-server architecture for connecting AI assistants to the places where your data lives , whether that's local files, databases, cloud services, or business applications. Think of it as a plugin system for AI agents: the agent discovers available tools, calls them during a conversation, and gets results back inline.
What it can convert
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. The full tool list covers a wide surface area: