# Markdownify MCP Server

A Model Context Protocol server that converts PDFs, images, audio, Office documents, and web pages, including YouTube transcripts, into Markdown text.

Markdownify MCP Server is a Model Context Protocol server that converts a range of file types and web content into Markdown. It handles PDFs, images, audio with transcription, DOCX, XLSX, and PPTX files, as well as web content such as YouTube video transcripts, Bing search results, and general web pages, and it can also retrieve existing Markdown files.

It is built with TypeScript and Bun, and its install step creates a Python virtual environment for the underlying conversion library it wraps. Path-based environment variables restrict which directories its file-input tools may read, guarding against reading files outside an allowed set of folders. It can be run directly, through Docker, or wired into a desktop AI application's own server configuration.

It is released under the MIT license.

Website: https://github.com/zcaceres/markdownify-mcp