File Format Converter (Pandoc)

🚀 MCP-Pandoc Documentation

MCP-Pandoc is a tool for file format conversion, supporting conversions between multiple formats. Users can convert text or files into formats such as PDF, Markdown, and HTML by specifying content or file paths.

🚀 Quick Start

MCP-Pandoc simplifies the process of file format conversion. With just a few steps, you can convert your files to the desired formats.

✨ Features

• Supports conversions between multiple file formats.
• Allows users to convert text or files by specifying content or file paths.

📦 Installation

Method 1: Manual Installation via Configuration File

• On MacOS, open the ~/Library/Application\ Support/Claude/claude_desktop_config.json file.
• On Windows, access the %APPDATA%/Claude/claude_desktop_config.json file.

"mcpServers": {
  "mcp-pandoc": {
    "command": "uv",
    "args": [
      "--directory",
      "<DIRECTORY>/mcp-pandoc",
      "run",
      "mcp-pandoc"
    ]
  }
}

"mcpServers": {
  "mcp-pandoc": {
    "command": "uvx",
    "args": [
      "mcp-pandoc"
    ]
  }
}

Method 2: Automatic Installation of Released Server Configuration via Smithery

Run the following bash command to automatically install the released MCP-Pandoc PyPI service using Smithery:

npx -y @smithery/cli install mcp-pandoc --client claude

⚠️ Important Note

To use the locally configured mcp-pandoc, refer to the "Development/Unreleased Server Configuration" section above.

📚 Documentation

Building and Publishing

To prepare the package for distribution:

1. Synchronize dependencies and update the lock file:

uv sync

2. Build the release version:

uv build

This will create source code and wheel format distribution packages in the dist/ directory.

3. Publish to PyPI:

uv publish

⚠️ Important Note

You need to set your PyPI credentials via environment variables or command-line parameters:

• Token: --token or UV_PUBLISH_TOKEN
• Or use username/password: --username/UV_PUBLISH_USERNAME and --password/UV_PUBLISH_PASSWORD

Debugging Guide

Since the MCP server runs through standard input and output, debugging can be challenging. For the best debugging experience, we highly recommend using MCP Inspector.

You can start MCP Inspector with the following command:

<Inspector 启动命令>

💻 Contribution

If you wish to contribute to the project, please refer to the Contribution Document to learn how to participate in development and submit code.