MarkItDown MCP Server

Name: MarkItDown MCP Server
Author: KorigamiK

@KorigamiK

リポジトリ

MarkItDown MCP Server について

A Model Context Protocol (MCP) server that converts various file formats to Markdown using the MarkItDown utility.

#python #mcp #mcp-server

基本情報

カテゴリ

その他

ライセンス

MIT

ランタイム

python

トランスポート

stdio

公開者

KorigamiK

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "markitdown_mcp_server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@KorigamiK/markitdown_mcp_server",
        "--client",
        "claude"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is MarkItDown MCP Server?

A Model Context Protocol (MCP) server that converts various file formats to Markdown using the MarkItDown utility. It is designed for use with any MCP-compliant client, such as Zed Editor or Claude Desktop.

How to use MarkItDown MCP Server?

Install via Smithery (npx -y @smithery/cli install @KorigamiK/markitdown_mcp_server --client claude) or manually (clone repo, run uv install). Configure your MCP client to run the server with uv run markitdown-mcp, then use the command /md <file> to convert a file to Markdown.

Key features of MarkItDown MCP Server

Converts PDF, PowerPoint, Word, and Excel files to Markdown
Extracts EXIF metadata and performs OCR on images
Transcribes speech and reads EXIF from audio files
Handles HTML, CSV, JSON, XML, and ZIP archives
Integrates with any MCP-compatible editor or tool

Use cases of MarkItDown MCP Server

Convert a PDF document to Markdown for note-taking or analysis
Extract text and metadata from images using OCR
Transcribe audio recordings and extract EXIF data
Process bulk office files (docx, xlsx, pptx) into unified Markdown
Unpack ZIP archives and convert all contained files in one step

FAQ from MarkItDown MCP Server

What file formats are supported?

PDF, PowerPoint, Word, Excel, Images (EXIF and OCR), Audio (EXIF and speech transcription), HTML, text-based formats (CSV, JSON, XML), and ZIP files (iterates over contents).