mcp-document-converter

The MCP Document Converter is a toolbox for transforming documents between Markdown, HTML, DOCX, PDF, and plain text formats. It exposes a conversion utility that can be invoked via MCP-compatible clients (such as Trae IDE, Claude Desktop, or Cherry Studio) to convert a source document from one format to a target format, while preserving metadata and supporting custom templates and styling when rendering outputs. Typical use involves installing the converter as an MCP service, then calling the convert_document tool with a source file and a target format (html, pdf, markdown, docx, or text). The converter can also accept options to tailor the output, such as CSS styling for HTML, or a specific rendering template for PDF or DOCX outputs. The server is designed to be easily extended with additional parsers and renderers through its plugin architecture, enabling future format support.

Within an MCP-enabled workflow, you can feed the converter a source document (via a file path or in-memory payload) and request a target format. The tool preserves essential metadata like title and author where possible and supports embedded assets and styling during conversion. You can also leverage the registry and conversion engine to automatically select compatible parsers and renderers for a given pair of formats, simplifying complex conversions in AI-assisted workflows.

Prerequisites:

• Python 3.10 or newer
• pip (Python package installer)
• Internet access to download dependencies

Installation steps:

1. Ensure Python is installed and accessible from the command line.
2. Install the document converter package from PyPI:

pip install mcp-document-converter

3. (Optional) If you prefer to install from source, clone the repository and install in editable mode:

git clone https://github.com/xt765/mcp-document-converter.git
cd mcp-document-converter
pip install -e .

4. Verify installation by listing the available command or running a quick help check (depending on how the package exposes CLI commands):

# Example, if the tool exposes a CLI entry point
mcp-document-converter --help

5. Configure MCP integration (example configuration is provided in the README):

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

Notes:

• If you are in a restricted network or China, you may prefer using the Gitee or GitHub paths as described in the README, or install via PyPI for a faster setup.
• The package can also be installed from source and wired into MCP using the provided configuration examples.

Tips and notes:

• Supported input formats: Markdown, HTML, DOCX, PDF, Text. Supported outputs: Markdown, HTML, DOCX, PDF, Text.
• When converting, you can customize output with options like templates, CSS, and metadata preservation. Check the API or documentation for the exact option keys.
• If your environment uses a proxy or firewall, ensure Python can reach PyPI for installation, or use a local mirror.
• For MCP integration, confirm your host environment can resolve the repository URL if you choose to install from Git sources.
• If you encounter encoding or font issues in PDF or DOCX outputs, verify the rendering engine and font availability on your system.
• Monitor resource usage when converting large documents, as images, tables, or embedded assets can impact memory and processing time.