XMind AI

XMind AI MCP is an integrated toolset that converts diverse formats to XMind, analyzes mind map structures with AI, and runs as an MCP server over UVX for fast, scriptable workflows. It enables batch conversions, AI-assisted structure optimization, and seamless integration with development environments.

How to use

You run the MCP server locally or in your development environment and connect to it from an MCP client or the standalone converter. Use the UVX deployment to start the server in fastMCP or STDIO mode, then submit conversion requests, analysis requests, or AI-assisted enhancements. The converter can read Markdown, HTML, Text, Word, Excel, and more, and output XMind files to a configured location.

Typical workflows you can perform include: converting a single document to an XMind map, batch converting a directory of resources, analyzing an existing XMind file for structure quality, and applying AI-driven improvements to the mind map layout. You can also integrate the server with Trae MCP workflow for automated pipelines.

How to install

Prerequisites you need before starting the MCP server: Node/UVX tooling for MCP deployment, Python, and a compatible shell environment.

Install and run the MCP server with UVX in standard mode, then verify the version and help output to confirm a successful setup.

Configuration and usage notes

Two MCP server configurations are provided for local development and standard deployment. The server uses UVX to launch the xmind-mcp process and can be customized with environment variables to ensure proper encoding and UTF handling.

{
  "mcpServers": {
    "xmind_mcp": {
      "command": "uvx",
      "args": ["xmind-mcp"],
      "env": {
        "PYTHONIOENCODING": "utf-8",
        "PYTHONUTF8": "1"
      },
      "description": "XMind MCP - UVX installed",
      "disabled": false,
      "autoApprove": []
    }
  }
}

Troubleshooting and tips

If you encounter encoding issues, ensure PYTHONIOENCODING is set to utf-8 and PYTHONUTF8 is enabled in the environment for the running MCP server. For development, you can run the local mode with explicit paths and check the output directory for generated XMind files.

Examples and workflows

Convert a Markdown file to XMind using the standalone converter and save the result to a designated output path, or rely on the server to place results in a configured absolute directory.

Paths & configuration

Configure a default absolute output directory for MCP tools via the server config or by using a CLI option. If no default is configured, pass an explicit absolute output_path/output_filepath in tool calls.

Available tools

read_xmind_file

Read and parse an existing XMind file into a structured representation for inspection or processing.

create_mind_map

Create a new mind map from a title and a JSON representation of topics, then write to an output path if provided.

analyze_mind_map

Analyze the structure and quality metrics of a given XMind file to assess organization and completeness.

convert_to_xmind

Convert input sources (Markdown, HTML, Word, etc.) into an XMind map and save to the specified output.

list_xmind_files

List XMind files present in a directory, with optional recursive search.

translate_xmind_titles

Translate mind map titles to a target language while preserving structure and hierarchy.