by YuChenSSR
Converts Markdown content into interactive mindmaps, offering HTML output or a saved file path for seamless integration with AI clients.
Converts Markdown files into interactive mindmaps using the markmap-cli
library. It can return the full HTML markup for immediate preview or save an HTML file and return its path, helping AI assistants reduce token usage.
pip install mindmap-mcp-server
or via uvx
. Docker is also supported.claude_desktop_config.json
(or any MCP‑compatible client). Example with uvx
:{
"mcpServers": {
"mindmap": {
"command": "uvx",
"args": ["mindmap-mcp-server", "--return-type", "html"]
}
}
}
Replace html
with filePath
to obtain a saved file.
ychen94/mindmap-converter-mcp
and run with a mounted volume. The container provides two tools: markdown-to-mindmap-content
(HTML) and markdown-to-mindmap-file
(saved file).filePath
mode to avoid sending large HTML payloads back to the AI client.python
or uvx
because the server calls markmap-cli
under the hood.uvx
or pip
method; the README recommends Docker for Windows due to potential npx
issues.--toolbar false
(or set toolbar: false
in the tool parameters) when invoking the tool.markdown-to-mindmap-file
mode, provide a filename
argument; otherwise a timestamped name is generated.--return-type filePath
do? It saves the generated HTML to the mounted output directory and returns the absolute path, allowing you to open it locally without extra token cost.A Model Context Protocol (MCP) server for converting Markdown content to interactive mindmaps.
pip install mindmap-mcp-server
Or using uvx
:
uvx mindmap-mcp-server
Or using docker
safer and easier.
Three installation methods have been successfully tested on macOS and Linux.
For Windows users experiencing issues with npx
for this MCP, consider using the Docker method. Alternatively, if you use Visual Studio Code, the "Markmap" extension offers a potentially simpler solution than navigating command-line tools.
If you're experiencing unresolved issues, you can use my recent system prompt as a Mindmap Assistant instead of using this MCP server.
You are a specialized assistant that generates HTML code for interactive markdown-based mind maps (markmaps). When a user sends you content, respond with a complete HTML document that displays their content as a markmap visualization.
If artifact tool is turned on, you can use the artifact.
Follow these requirements:
1. Use the markmap-autoloader library (version 0.18 or latest stable version)
2. Format the HTML exactly according to the template below
3. Replace the demo content in the template with the user's content, preserving their hierarchical structure
4. Maintain the markmap configuration options (maxWidth: 300, colorFreezeLevel: 2)
5. If the user doesn't provide markdown formatting (# for headings), format their content appropriately with main topics using # and subtopics using ##
Template to follow:
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta http-equiv="X-UA-Compatible" content="IE=edge" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Markmap</title>
<style>
svg.markmap {
width: 100%;
height: 100vh;
}
</style>
<script src="https://cdn.jsdelivr.net/npm/markmap-autoloader@0.18"></script>
</head>
<body>
<div class="markmap">
<script type="text/template">
---
markmap:
maxWidth: 300
colorFreezeLevel: 2
---
# markmap
## Links
- <https://markmap.js.org/>
- [GitHub](https://github.com/markmap/markmap)
## Related
- [coc-markmap](https://github.com/markmap/coc-markmap)
- [gatsby-remark-markmap](https://github.com/markmap/gatsby-remark-markmap)
## Features
- links
- **inline** ~~text~~ *styles*
- multiline
text
- `inline code`
- Katex - $x = {-b \pm \sqrt{b^2-4ac} \over 2a}$
- This is a very very very very very very very very very very very very very very very long line.
</script>
</div>
</body>
</html>
Visualization options: (If formulas or symbols don’t display correctly, download the HTML file and open it in a browser.)
View the mindmap in artifacts (if available):
Render the HTML file as a mindmap:
This package requires Node.js to be installed when using command python
or uvx
to run the server.
Add this server to your claude_desktop_config.json
:
{
"mcpServers": {
"mindmap": {
"command": "uvx",
"args": ["mindmap-mcp-server", "--return-type", "html"]
}
}
}
or
recommended:
{
"mcpServers": {
"mindmap": {
"command": "uvx",
"args": ["mindmap-mcp-server", "--return-type", "filePath"]
}
}
}
we use --return-type
to specify the return type of the mindmap content, you can choose html
or filePath
according to your needs.
html
will return the entire HTML content of the mindmap, which you can preview in your AI client's artifact;
filePath
will save the mindmap to a file and return the file path,which you can open in your browser. It can save your tokens !
Using a specific Python file in this repository:
{
"mcpServers": {
"mindmap": {
"command": "python",
"args": ["/path/to/your/mindmap_mcp_server/server.py", "--return-type", "html"]
}
}
}
or
{
"mcpServers": {
"mindmap": {
"command": "python",
"args": ["/path/to/your/mindmap_mcp_server/server.py", "--return-type", "filePath"]
}
}
}
we use --return-type
to specify the return type of the mindmap content, you can choose html
or filePath
according to your needs. see using `uvx` for more details.
First, you pull the image:
docker pull ychen94/mindmap-converter-mcp
Second, set the server:
{
"mcpServers": {
"mindmap-converter": {
"command": "docker",
"args": ["run", "--rm", "-i", "-v", "/path/to/output/folder:/output", "ychen94/mindmap-converter-mcp:latest"]
}
}
}
⚠️ Replace /path/to/output/folder
with an actual path on your system where you want to save mind maps, such as /Users/username/Downloads
on macOS or C:\\Users\\username\\Downloads
on Windows.
Tools Provided in the docker container The server provides the following MCP tools:
-v
and /path/to/output/folder:/output
in the command docker
.File Not Found
If your mind map file isn't accessible:
1 Check that you've correctly mounted a volume to the Docker container
2 Ensure the path format is correct for your operating system
3 Make sure Docker has permission to access the directory
Docker Command Not Found
1 Verify Docker is installed and in your PATH
2 Try using the absolute path to Docker
Server Not Appearing in Claude
1 Restart Claude for Desktop after configuration changes
2 Check Claude logs for connection errors
3 Verify Docker is running
Advanced Usage
Using with Other MCP Clients
This server works with any MCP-compatible client, not just Claude for Desktop. The server implements the Model Context Protocol (MCP) version 1.0 specification.
This server provides a tool for converting Markdown content to mindmaps using the markmap-cli
library:
In Claude, you can ask:
"give a mindmap for the following markdown code, using a mindmap tool:
# Project Planning
## Research
### Market Analysis
### Competitor Review
## Design
### Wireframes
### Mockups
## Development
### Frontend
### Backend
## Testing
### Unit Tests
### User Testing
"
if you want to save the mindmap to a file, and then open it in your browser using the iTerm MCP server:
"give a mindmap for the following markdown input_code using a mindmap tool, after that,use iterm to open the generated html file. input_code:
markdown content
"
"Think about the process of putting an elephant into a refrigerator, and provide a mind map. Open it with a terminal."
and more
This project is licensed under the MIT License. For more details, please see the LICENSE file in this project repository
If this project is helpful to you, please consider giving it a Star ⭐️
The advancement of technology ought to benefit all individuals rather than exploit the general populace.
Please log in to share your review and rating for this MCP.
{ "mcpServers": { "mindmap-mcp-server": { "command": "npx", "args": [ "-y", "mindmap-mcp-server" ], "env": {} } } }
Explore related MCPs that share similar capabilities and solve comparable challenges
by zed-industries
A high‑performance, multiplayer code editor designed for speed and collaboration.
by modelcontextprotocol
Model Context Protocol Servers
by modelcontextprotocol
A Model Context Protocol server for Git repository interaction and automation.
by modelcontextprotocol
A Model Context Protocol server that provides time and timezone conversion capabilities.
by cline
An autonomous coding assistant that can create and edit files, execute terminal commands, and interact with a browser directly from your IDE, operating step‑by‑step with explicit user permission.
by continuedev
Enables faster shipping of code by integrating continuous AI agents across IDEs, terminals, and CI pipelines, offering chat, edit, autocomplete, and customizable agent workflows.
by upstash
Provides up-to-date, version‑specific library documentation and code examples directly inside LLM prompts, eliminating outdated information and hallucinated APIs.
by github
Connects AI tools directly to GitHub, enabling natural‑language interactions for repository browsing, issue and pull‑request management, CI/CD monitoring, code‑security analysis, and team collaboration.
by daytonaio
Provides a secure, elastic infrastructure that creates isolated sandboxes for running AI‑generated code with sub‑90 ms startup, unlimited persistence, and OCI/Docker compatibility.