Xmind Generator MCP Server
About
An MCP (Model Context Protocol) server for generating Xmind mind maps. This server allows LLMs to create structured mind maps through the MCP protocol.
Details
- Author
- BangyiZhang
- GitHub stars
- 44
- Downloads
- 350
- Categories
- Productivity, Other, Automation, Developer Tools
Jump to
- Generate hierarchical Xmind mind maps with topics
- Support topic notes, labels, markers, and relationships
- Save generated mind maps to local files
- Read existing .xmind files and export as Markdown outline
- Integrates with Claude Desktop and MCP clients
- Requires Node.js 18+ and Xmind desktop application
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Xmind Generator MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Install via npx (recommended) or clone and build locally. Configure the Claude Desktop configuration file (claude_desktop_config.json) with the MCP server command and optional environment variables (outputPath, autoOpenFile). Use the generate-mind-map tool to create a mind map from a hierarchical topic structure, or the read-mind-map tool to export an existing .xmind file as a Markdown outline.
generate-mind-map
read-mind-map
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"xmind generator mcp server": {
"xmind-generator": {
"command": "npx",
"args": [
"xmind-generator-mcp"
],
"env": {
"outputPath": "/path/to/save/xmind/files",
"autoOpenFile": "false"
}
}
}
}
}
McpServers
{
"xmind-generator": {
"command": "npx",
"args": [
"xmind-generator-mcp"
],
"env": {
"outputPath": "/path/to/save/xmind/files",
"autoOpenFile": "false"
}
}
}
Xmind Generator MCP Server
An MCP (Model Context Protocol) server for generating Xmind mind maps. This server allows LLMs to create structured mind maps through the MCP protocol.
Features
- Generate Xmind mind maps with hierarchical topic structures
- Support for topic notes, labels, markers, and relationships
- Save mind maps to local files
- Read existing .xmind files and export an outline as Markdown
- Easy integration with Claude Desktop and other MCP clients
Prerequisites
- Node.js: Version 18 or higher is required
- Xmind: Install Xmind desktop application to open and edit the generated mind maps
- Claude Desktop: Required to use this tool as an extension
Setup with Claude Desktop
Option 1: Using npx (Recommended)
1. Create or edit the Claude Desktop configuration file:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
2. Add the following configuration:
{
"mcpServers": {
"xmind-generator": {
"command": "npx",
"args": ["xmind-generator-mcp"],
"env": {
"outputPath": "/path/to/save/xmind/files",
"autoOpenFile": "false"
}
}
}
}
3. Restart Claude Desktop
4. Start using the Xmind generator in your conversations
Option 2: Local Installation
1. Clone the repository:
git clone https://github.com/BangyiZhang/xmind-generator-mcp.git
cd xmind-generator-mcp
npm install
npm run build
2. Create or edit the Claude Desktop configuration file:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
3. Add the following configuration:
{
"mcpServers": {
"xmind-generator": {
"command": "node",
"args": ["path/to/xmind-generator-mcp/dist/index.js"],
"env": {
"outputPath": "/path/to/save/xmind/files",
"autoOpenFile": "false"
}
}
}
}
4. Replace path/to/xmind-generator-mcp with the actual path to your cloned project
5. Restart Claude Desktop
6. Start using the Xmind generator in your conversations
Note: The env section is optional. It allows you to set environment variables for the server:
- outputPath: Default directory or file path where Xmind files will be saved. This can be overridden by the outputPath parameter in the tool call.
- autoOpenFile: Controls whether generated Xmind files are automatically opened after creation. Set to "false" to disable auto-opening (default is "true").
Available Tools
generate-mind-map
Generates an Xmind mind map from a hierarchical structure of topics.
Parameters:
- title (string): The title of the mind map (root topic)
- topics (array): Array of topics to include in the mind map
- title (string): The title of the topic
- ref (string, optional): Reference ID for the topic
- note (string, optional): Note for the topic
- labels (array of strings, optional): Labels for the topic
- markers (array of strings, optional): Markers for the topic (format: "Category.name", e.g., "Arrow.refresh")
- children (array, optional): Array of child topics
- relationships (array, optional): Array of relationships between topics
- outputPath (string, optional): Custom output path for the Xmind file. This overrides the environment variable if set.
read-mind-map
Reads an existing .xmind file and exports it as a Markdown outline.
Parameters:
- inputPath (string): Path to the .xmind file
- style (string, optional): Markdown style. Currently supports A (outline). (B reserved for future richer exports.)
Example:
{
"inputPath": "/path/to/file.xmind",
"style": "A"
}
Example
Here's an example of how to use the generate-mind-map tool:
{
"title": "Project Plan",
"topics": [
{
"title": "Research",
"ref": "topic:research",
"note": "Gather information about the market",
"children": [
{
"title": "Market Analysis",
"labels": ["Priority: High"]
},
{
"title": "Competitor Research",
"markers": ["Task.quarter"]
}
]
},
{
"title": "Development",
"children": [
{
"title": "Frontend",
"markers": ["Arrow.refresh"]
},
{
"title": "Backend"
}
]
}
]
}
License
MIT
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.


