# Markdownify

Markdownify transforms PDFs, office documents, audio recordings, and web pages into clean Markdown. It lets coding assistants inspect varied media files directly.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: zcaceres
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/zcaceres/markdownify-mcp
- Updated: 2026-10-29
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/markdownify

## What it does

This MCP server processes raw files and online content, returning plain Markdown text to the assistant. Supported inputs include PDF papers, DOCX files, Excel sheets, presentations, and OCR for images. It can also pull transcripts from YouTube clips and convert search results.

Under the hood, the server relies on a local Python environment running `markitdown`. It builds with Bun, compiling TypeScript into a Node runtime bundle. To protect your filesystem, define `MD_ALLOWED_PATHS`. This variable prevents tools from reading directories outside your designated list.

The published Docker container provides only `markitdown[pdf]`. Sound transcription and image recognition will fail in the container unless you run the full stack locally with Bun.

## Requirements
- Needs: Docker

## Install

### Claude Code (native platform)

**Install**
- The MCP registry has no launch command for it: see the repository README

## When to use it
- turn spreadsheets and slide decks into readable Markdown tables
- transcribe speech from sound files directly into context
- extract clean text and structure from PDF or DOCX files
- pull transcripts from YouTube links for summarization
- restrict agent file access to specific allowed folders

## Example prompts
- Convert the spreadsheet `data.xlsx` into Markdown format.
- Extract the full transcript from this YouTube video URL.
- Transcribe the voice memo `audio.m4a` and save it as Markdown.

## Similar
- [markdown-converter](https://affpapa.org/en/nevibecode/skills/markdown-converter.md) — skill · 1700 installs
- [convert-pdf-to-md](https://affpapa.org/en/nevibecode/skills/convert-pdf-to-md.md) — skill · 138 installs
- [convert-word-to-md](https://affpapa.org/en/nevibecode/skills/convert-word-to-md.md) — skill · 85 installs
- [markitdown](https://affpapa.org/en/nevibecode/skills/jimmc414-markitdown.md) — skill · 1 install
- [equalang-mcp](https://affpapa.org/en/nevibecode/mcp/equalang-mcp.md) — MCP server · ★ 22 repo stars
- [librarian](https://affpapa.org/en/nevibecode/skills/librarian.md) — skill · ★ 38 repo stars

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
