M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
M

Markdownify MCP Server

Community
Convert diverse files and web content to Markdown.
How to install ↓ Compare GitHub source repository ↗
Category
Other #150 of 230
Stars
★ 3.0k Very popular
Transport
stdio (local process)
Runtime
Bun · Node.js · Python
Credentials
No credential needed
License
MIT
Last commit
Tools
10
50FMRS · D

This MIT-licensed community MCP server focuses on converting diverse files, web content, and selected media content into Markdown. The README documents installation, startup, desktop-app configuration, and path restrictions, but does not provide a verified list of specific compatible clients or a server manifest.

Strongest · Documentation 13/20 Weakest · Maintenance 7/20

Reliability
8/20
Security and permissions
11/20
Maintenance
7/20
Documentation
13/20
Setup experience
11/20
Why each score
Reliability 8/20
The README provides Bun installation, build, startup steps, and a tool list, and identifies dependencies such as MarkItDown and Repomix, so the normal path is plausible. However, there is no server manifest, source code, CI, committed tests, or handshake evidence. Python virtual-environment and Docker image differences also create dependency uncertainty, so the declared tools cannot be verified against actual behavior and points are deducted.
Security and permissions 11/20
The described tools primarily convert files and web content; no deletion, payment, deployment, or remote-command capability is shown, and no credentials are requested, so no security red line is evident. The web, YouTube, and Bing tools imply external network access, while local file access is unrestricted by default and MD_ALLOWED_PATHS is optional. There is no source evidence for path-validation details, input limits, network scoping, or data-flow handling, so points are deducted.
Maintenance 7/20
The repository is not archived and declares an MIT license. The issue count is only a discovery signal and does not establish maintenance quality. No release history, commit cadence, issue-response record, dependency-update process, security-response channel, or clear maintainer information is provided, leaving sustained maintenance weakly evidenced.
Documentation 13/20
The README covers installation, build and startup, desktop-client configuration, tool names, environment variables, Docker mounts, and the slim-image feature limitation, making it reasonably usable. It does not document per-tool parameters, return formats, authentication, resource or cost limits, error behavior, or system-level troubleshooting. The manifest is also unavailable, so a high score is not justified.
Setup experience 11/20
The README gives a path from Bun installation through build and startup, plus a mainstream desktop-client configuration example and a Docker option. The setup still depends on Bun, Node configuration, a Python virtual environment, and external executables; platform differences and dependency-failure handling are not verified, and no CI or tests demonstrate a successful connection. It is plausible but manual and fragile.

Static review · not runListed 2026-07-29

Read the FMRS scoring method →

Fit and risk

What it can accessReads local filesUses the network

Best for

  • Users who need to normalize multiple file types and web content into Markdown.
  • Users who want local MCP desktop-app integration for Markdown conversion.
  • Developers comfortable configuring Bun, Node.js, and Python dependencies.

Not for

  • Users seeking structured database or cloud-storage operations.
  • Users requiring verified compatibility with a specific MCP client.
  • Users running the slim Docker image who need audio transcription or image OCR.

Required permissions

  • The server needs to read the file paths supplied to its file-input tools; MD_ALLOWED_PATHS can restrict readable directories.
  • When MD_ALLOWED_PATHS is unset, file access is unrestricted by that setting.
  • Tools handling web pages, YouTube, or Bing content need access to the relevant web content; the source does not state any additional authentication requirements.

Risks and side effects

  • The default file-reading scope is unrestricted; set MD_ALLOWED_PATHS to narrow the readable boundary.
  • The published Docker image installs only markitdown[pdf], so audio transcription and image OCR tools fail in that slim image.
  • When using Docker, mount only required directories and pass container paths to the tools.

Setup

Before you start

Runtime：Bun · Node.js · Python

Other optional settings (4)
MARKITDOWN_PATH optional Absolute path to the markitdown executable; set when markitdown is not in the bundled .venv or on PATH.
REPOMIX_PATH optional Absolute path to the repomix executable used by git-repo-to-markdown; set when not on PATH.
MD_ALLOWED_PATHS optional Colon-separated (semicolon on Windows) list of directories that all file-input tools are allowed to read.
MD_SHARE_DIR optional Deprecated single-directory alias for MD_ALLOWED_PATHS, kept for backward compatibility.

Clone the repository, run bun install to install dependencies, run bun run build to build the project, and run bun start to start the server. For a desktop app, set command to node and provide the absolute path to dist/index.js in args.

MCP config · JSON 
{"mcpServers":{"markdownify":{"command":"node","args":["{ABSOLUTE PATH TO FILE HERE}/dist/index.js"]}}}

Works in clients that read an "mcpServers" key. For clients such as VS Code that use a different key, the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "markdownify": {
      "command": "node",
      "args": [
        "{ABSOLUTE PATH TO FILE HERE}/dist/index.js"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add markdownify -- node '{ABSOLUTE PATH TO FILE HERE}/dist/index.js'

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

The client's tool list should show tools like youtube-to-markdown, pdf-to-markdown, and get-markdown-file; try 'use bing-search-to-markdown to search for Model Context Protocol' to confirm the connection works.

Troubleshooting

  1. Confirm that bun install and bun run build have completed.
  2. Confirm that the server is started with bun start and that the client points to the absolute path of dist/index.js.
  3. If markitdown is not in the bundled virtual environment, check MARKITDOWN_PATH or the system PATH.
  4. If git-repo-to-markdown functionality has issues, check REPOMIX_PATH or the system PATH.
  5. In Docker, verify that tools receive container paths and that MD_ALLOWED_PATHS includes the mounted directory.
  6. For audio transcription or image OCR, use the full local installation rather than the PDF-only Docker image.

Things to try

Once connected, you can ask your AI assistant things like:

  • Convert the YouTube video transcript to Markdown
  • Convert report.pdf on my desktop to Markdown
  • Search Bing for 'Model Context Protocol' and convert the results to Markdown
  • Fetch /data/notes.md and show me its contents

Tools 10

youtube-to-markdown read-only
Convert YouTube videos to Markdown.
pdf-to-markdown read-only
Convert PDF files to Markdown.
bing-search-to-markdown read-only
Convert Bing search results to Markdown.
webpage-to-markdown read-only
Convert web pages to Markdown.
image-to-markdown read-only
Convert images to Markdown with metadata.
audio-to-markdown read-only
Transcribe audio files and convert them to Markdown.
docx-to-markdown read-only
Convert DOCX files to Markdown.
xlsx-to-markdown read-only
Convert XLSX files to Markdown.
Show 2 more tools
pptx-to-markdown read-only
Convert PPTX files to Markdown.
get-markdown-file read-only
Retrieve an existing Markdown file; the extension must be .md or .markdown.

Use cases

Convert office documents, PDFs, presentations, and spreadsheets into readable Markdown.
Extract content from images and audio, including audio transcription.
Organize web pages, YouTube transcripts, or Bing search results as Markdown.
Retrieve existing Markdown files.

Overview

Markdownify is a Model Context Protocol (MCP) server that converts PDFs, images, audio, DOCX, XLSX, PPTX, web pages, selected search results, and YouTube content into Markdown. It can also retrieve existing Markdown files.

Similar servers

BioMCP 75 · B

One binary. One grammar. Evidence from the biomedical sources you already trust.

★ 659 · Tools 11 Compare with this →
Compare with other MCP servers

Source revision 024f97cea9a9 Data synced 2026-10-19 Read the FMRS scoring method