mcp-pymupdf

An MCP server for reading PDFs using PyMuPDF. Supports local files and URLs.

Installation

Terminal

pip install mcp-pymupdf

Extract text from a PDF file or URL.

config.json

{
  "source": "/path/to/file.pdf",
  "max_pages": 50,
  "max_bytes": 200000
}

Extract text from a specific page range.

config.json

{
  "source": "https://example.com/paper.pdf",
  "start_page": 3,
  "end_page": 8
}

Retrieve PDF metadata without reading full text.

config.json

{
  "source": "/path/to/file.pdf"
}

Add to your Claude Desktop config:

config.json

{
  "mcpServers": {
    "pdf": {
      "command": "mcp-pymupdf"
    }
  }
}

MIT