MarkltDown MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MarkltDown MCP Serverconvert my quarterly_report.pdf to markdown"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MarkltDown — File → Markdown (riduci i token in Claude)
Converte PDF, Word, Excel, PowerPoint, immagini (OCR), CSV/JSON/XML/HTML, EPub, audio e ZIP in Markdown, usando markitdown di Microsoft.
Perché: dare a Claude un .md leggero invece del file originale (PDF/immagini)
riduce molto i token. Converti fuori dalla chat e passa solo il Markdown.
Dove finiscono i Markdown
File dentro un progetto in
~/solongevity-projects/<progetto>/…→ tutti raccolti in<progetto>/_markdown/(struttura interna preservata)File altrove (es. Download) → cartella
_markdown/accanto al file
Related MCP server: DocMistral MCP Server
Struttura del repo
markletdown-mcp/
├── README.md
├── QUICK_START.md # guida per l'MCP server (Claude Desktop)
├── requirements.txt
├── server/
│ └── file_to_markdown_mcp.py # MCP server (file→md per path dentro Claude Desktop)
├── macos/
│ ├── install.command # ⟵ doppio click: installa tutto
│ ├── convert.command # motore di conversione
│ ├── app.applescript # sorgente dell'app droplet sul Desktop
│ ├── quickaction/ # sorgente della Quick Action (tasto destro Finder)
│ └── icon/ # icona (make_icon.py, AppIcon.icns, icon.png)
└── windows/
├── install.ps1 # ⟵ installa venv + voce tasto destro
├── convert.ps1 # motore di conversione
├── convert.bat # doppio click
└── icon/AppIcon.ico
.venv/viene creato dall'installer in ciascuna macchina e NON è versionato.
Installazione
macOS
Prerequisiti: brew install python@3.12 pandoc tesseract
git clone https://github.com/guido1967-lab/markletdown-mcp.git
cd markletdown-mcp
# Doppio click su macos/install.command (oppure:)
./macos/install.commandOttieni: app "Converti in Markdown" sul Desktop + Quick Action nel Finder (tasto destro su file/cartelle → Azioni rapide → Converti in Markdown).
Windows
Prerequisiti: Python 3.10+ (da python.org, "Add to PATH"); consigliato winget install JohnMacFarlane.Pandoc.
git clone https://github.com/guido1967-lab/markletdown-mcp.git
cd markletdown-mcp\windows
powershell -ExecutionPolicy Bypass -File .\install.ps1Ottieni: voce "Converti in Markdown" nel menu tasto-destro di file e cartelle
(+ convert.bat per il doppio click). Per rimuovere: .\install.ps1 -Uninstall.
MCP server (opzionale, Claude Desktop)
Per convertire file su disco dall'interno di Claude Desktop, vedi
QUICK_START.md. Per il puro risparmio di token è preferibile
pre-convertire su disco (app/Quick Action) e passare solo i .md.
This server cannot be deployed
Maintenance
Related MCP Connectors
Convert documents and web pages to clean Markdown: PDF, DOCX, XLSX, EPUB, scanned files, any URL.
Convert PDF, DOCX, HTML, and URLs to clean, LLM-ready markdown with tables preserved
Convert files, URLs, and documents to clean, AI-ready Markdown via MCP.
Parse PDF/Word/PPT/HTML to Markdown; tables as JSON, image extraction, RAG chunking, page ranges.
Related MCP Servers
- FlicenseAqualityDmaintenanceEnables document conversion between PDF, DOCX, and Markdown formats to facilitate reading and editing complex files in AI tools like Claude Desktop or Cursor. It utilizes marker-pdf and pandoc to provide structured text versions of documents, helping to manage context and support unsupported file types.11-
- AlicenseNot gradedqualityAmaintenanceConverts documents and images to Markdown using Mistral AI's OCR, enabling AI-powered document processing via MCP-compatible clients like Claude Desktop.33 npm2MIT
- AlicenseAqualityBmaintenanceTransform any local file into AI-optimized markdown format for seamless integration with Claude Desktop, Claude Code, and other MCP clients.21MIT
- AlicenseAqualityBmaintenanceAutomatically converts PDF, DOCX, XLSX, and CSV files to clean markdown when read by Claude Code, reducing token usage by up to 98%.17 npmMIT