Skip to main content
Glama
robertZaufall

MindManager MCP Server

README.md
> [!WARNING]
> **Deprecated:** This MCP will not be developed further. Please use the Codex/Claude skills from https://github.com/robertZaufall/mindm-skill instead.

# MindManager MCP Server

A Model Context Protocol (MCP) server implementation for the `mindm` library, providing a standardized interface to interact with MindManager on Windows and macOS.  

[![PyPI version](https://img.shields.io/pypi/v/mindm-mcp.svg?label=pypi%20(mindm-mcp)&style=for-the-badge&color=blue)](https://pypi.org/project/mindm-mcp/)
[![PyPI version](https://img.shields.io/pypi/v/mindm.svg?label=pypi%20(mindm)&style=for-the-badge&color=lightblue)](https://pypi.org/project/mindm/)

## Overview

This server allows you to programmatically interact with MindManager through the Model Context Protocol (MCP), a standardized way to provide context and tools to LLMs. It leverages the `mindm` library to manipulate MindManager documents, topics, relationships, and other mindmap elements.

### Animated examples (using Codex in VSCode on Windows)

- Example
Refining a single topic including tags, icons for the new topics: 

_get the current Mindmanager mindmap as mermaid (full), refine the given topics at least two levels deep (not just topics), add a tag "Important" to the 3 top most important topics and add corresponding stock icons to every topic but the central topic and create the new MindManager mindmap_

![MindManager MCP in Codex](assets/refine_map_anim2.gif) 
 
- Example
Refining a single topic including generating text in notes for the new topics: 
![MindManager MCP in Codex](assets/refine_map_anim.gif) 

- Example
Cloning a map including all supported topic properties:
![MindManager MCP in Codex](assets/clone_map_anim.gif) 
 
## Other examples

### Client

`get the current MindManager mindmap, translate every topic to German and create a new MindManager mindmap (simple).` 
 
`get the current MindManager mindmap, refine each topic 2 levels and create a new MindManager mindmap (simple).` 

### Shell:

`codex exec "get the current MindManager mindmap, add a meaningful emoji to every topic and create new MindManager mindmap (simple)"` 

_remark: emojis only work on macOS._ 

## Features

- Retrieve mindmap structure and central topics
- Export mindmaps to Mermaid, Markdown, JSON formats to be used in LLM chats
- Create MindManager mindmaps directly from Mermaid (full or simplified syntax)
- Fetch MindManager and package versions for quick diagnostics
- Get information about MindManager installation and library folders
- Get current selection from MindManager

## Planned Features

- Add, modify, and manipulate topics and subtopics without Mermaid roundtrips
- Add relationships between topics
- Add tags to topics
- Set document background images

## Requirements

- Python 3.12 or higher
- `mcp` package (Model Context Protocol SDK)
- `mindm` library (included in this project)
- MindManager (supported versions: 23-) installed on Windows or macOS

## Installation macOS

```bash
# Clone the repository (if you're using it from a repository)
git clone https://github.com/robertZaufall/mindm-mcp.git
cd mindm-mcp

# create a virtual environment for Python
brew install uv # if needed
uv pip install -r pyproject.toml

# alternative: manual installation of modules
uv add "mcp[cli]"
uv add fastmcp
uv add markdown-it-py
uv add -U --index-url=https://test.pypi.org/simple/ --extra-index-url=https://pypi.org/simple/ mindm mindm-mcp
```

## Installation Windows

```bash
# Change to DOS command prompt
cmd

# Clone the repository (if you're using it from a repository)
git clone https://github.com/robertZaufall/mindm-mcp.git
cd mindm-mcp

# create a virtual environment for Python
pip install uv # if needed
uv pip install -r pyproject.toml

# install nodejs
choco install nodejs # if you have chocolatey installed. If not install nodejs otherwise
refreshenv
node -v
npm install -g npx
```

## Usage

### MCP inspector

```bash
# run mcp with inspector
uv run --with mind --with fastmcp --with markdown-it-py mcp dev mindm_mcp/server.py
```

### Claude Desktop

#### Local python file

Adjust the path for the local file as needed.
```json
{
  "mcpServers": {
    "mindm (MindManager)": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mindm>=0.0.5.3",
        "--with",
        "fastmcp",
        "--with",
        "markdown-it-py",
        "/Users/master/git/mindm-mcp/mindm_mcp/server.py"
      ]
    }
  }
}
```

#### Module from package repository

Adjust `VIRTUAL_ENV` as needed.
```json
{
    "mcpServers": {
      "mindm (MindManager)": {
        "command": "uv",
        "args": [
          "run",
          "--with",
          "mindm>=0.0.5.3",
          "--with",
          "mindm-mcp>=0.0.2.1",
          "--with",
          "fastmcp",
          "--with",
          "markdown-it-py",
          "-m",
          "mindm_mcp.server"
        ],
        "env": {
            "VIRTUAL_ENV": "/Users/master/git/mindm-mcp/.venv"
        }
      }
  }
}
```

Hint: If the MCP server does not show up with the hammer icon on Windows, close Claude Desktop and kill all background processes.  

### Codex (VSCode Extension / CLI)

#### Local python file

`config.toml` (adjust the path for the local file as needed).
```
[features]
rmcp_client = true

[mcp_servers.mindmanager]
command = "uv"
args = ["run", "--with", "mindm>=0.0.5.3", "--with", "fastmcp", "--with", "markdown-it-py", "/Users/master/git/mindm-mcp/mindm_mcp/server.py"]

```

#### Module from package repository

`config.toml` 
```
[features]
rmcp_client = true

[mcp_servers.mindmanager]
command = "uv"
args = ["run", "--with", "mindm>=0.0.5.3", "--with", "fastmcp", "--with", "markdown-it-py", "--with", "mindm-mcp>=0.0.2.1", "-m", "mindm_mcp.server"]

```

### VSCode Chat (GitHub Copilot)

#### Local python file:

Adjust the path for the local file as needed.
```
uv run --with mindm>=0.0.5.3 --with fastmcp --with markdown-it-py /Users/master/git/mindm-mcp/mindm_mcp/server.py
```
 
or server definition in `mcp.json`:
```
		"Mindmanager": {
			"type": "stdio",
			"command": "uv",
			"args": [
				"run",
				"--with",
				"mindm>=0.0.5.3",
				"--with",
				"fastmcp",
				"--with",
				"markdown-it-py",
				"/Users/master/git/mindm-mcp/mindm_mcp/server.py"
			]
		}
```

#### Module from package repository

```
uv run --with mindm>=0.0.5.3 --with fastmcp --with markdown-it-py --with mindm-mcp>=0.0.2.1 -m mindm_mcp.server
```

or server definition in `mcp.json`:
```
		"Mindmanager": {
			"type": "stdio",
			"command": "uv",
			"args": [
				"run",
				"--with",
				"mindm>=0.0.5.3",
				"--with",
				"fastmcp",
				"--with",
				"markdown-it-py",
				"--with",
				"mindm-mcp>=0.0.2.1",
				"-m",
				"mindm_mcp.server"
			]
		}
```

## MCP Tools

The server exposes the following tools through the Model Context Protocol:

### Document Interaction
- `get_mindmap`: Retrieves the current mindmap structure from MindManager
- `get_selection`: Retrieves the currently selected topics in MindManager
- `get_library_folder`: Gets the path to the MindManager library folder
- `get_mindmanager_version`: Gets the installed MindManager version
- `get_grounding_information`: Extracts grounding information (central topic, selected subtopics) from the mindmap

### Serialization
- `serialize_current_mindmap_to_mermaid`: Serializes the currently loaded mindmap to Mermaid format
- `serialize_current_mindmap_to_markdown`: Serializes the currently loaded mindmap to Markdown format
- `serialize_current_mindmap_to_json`: Serializes the currently loaded mindmap to a detailed JSON object with ID mapping

### Creation
- `create_mindmap_from_mermaid`: Build a MindManager map from Mermaid (full syntax with IDs and metadata)
- `create_mindmap_from_mermaid_simple`: Build a MindManager map from simplified Mermaid text

### Versioning
- `get_versions`: Returns the `mindm-mcp` and `mindm` package versions for debugging

## Platform Support

- **Windows**: Full support for topics, notes, icons, images, tags, links, relationships, and RTF formatting
- **macOS**: Support for topics, notes, and relationships (limited support compared to Windows)

## Integration with Claude and other LLMs

This MCP server can be installed in Claude Desktop or other MCP-compatible applications, allowing LLMs to:

1. Access mindmap content
2. Manipulate mindmaps (coming)
3. Create new mindmaps based on LLM-generated content (coming)

## Troubleshooting

- Ensure MindManager is running before starting the server
- For macOS, make sure you allow Claude Desktop to automate MindManager

## MCPHub

[Certified on MCPHub](https://mcphub.com/mcp-servers/robertZaufall/mindm-mcp)

## Acknowledgements

This project is built upon the `mindm` library, providing Python interfaces to MindManager on Windows and macOS platforms. It uses the Model Context Protocol (MCP) SDK developed by Anthropic.

## License

MIT License - See LICENSE file for details

TDQS

A3.7/5.0

Scored across 11 tools

Disambiguation4/5

Most tools have distinct purposes, but there is some overlap between 'get_mindmap' and 'serialize_current_mindmap_to_json' as both retrieve mindmap data in structured formats. The three serialization tools (to JSON, Markdown, Mermaid) are clearly differentiated by output format, and creation vs. retrieval tools are well separated.

Naming Consistency5/5

All tools follow a consistent snake_case verb_noun pattern with clear action prefixes (create_, get_, serialize_). The naming is highly predictable, with 'serialize_current_mindmap_to_*' showing excellent consistency for the serialization family.

Tool Count5/5

11 tools is well-scoped for a MindManager automation server. The set covers creation (2 tools), retrieval (4 tools), serialization (3 tools), and utility functions (2 tools) without being overwhelming. Each tool appears to serve a distinct role in the mindmap workflow.

Completeness4/5

The server provides comprehensive coverage for mindmap creation, retrieval, and serialization workflows. Minor gaps exist in update/delete operations for mindmap content, but the core functionality for importing, exporting, and inspecting mindmaps is well covered. The inclusion of both simple and advanced creation tools shows thoughtful design.

Maintenance

ActivityInactive
ResponsivenessNo issues