MCP-Mind Map Server

A dynamic MCP server management service that creates, runs, and manages Model Context Protocol (MCP) servers dynamically. This service serves as an MCP server and launches/manages other MCP servers as child processes, enabling a flexible MCP ecosystem.

Index

• Introduction

• Project structure

• Technologies used

• Requirements

• How to install on Claude Desktop

• Useful links

• Contributions

• License

• Contact

Related MCP server: CaptureMind

Introduction

The mapas_mentais project is a Python application that generates automated mind maps to facilitate the study, review, comparison and presentation of different topics. Using the idea of MCP-server, the system provides insights by interacting directly with Claude Desktop through Claude models. Ideal for students, teachers and professionals who want to organize ideas visually and efficiently, the project is easily extensible and can be integrated with other automation systems or virtual assistants.

Project structure

The idea for this project came from the explanations given by Professor Sandeco Macedo, from UFG (Federal University of Goiás), about MCPs through the book MCP and A2A for Dummies . It is a simple MCP-Server that uses only the FastMCP package, also following the guidelines of the official repository of the Model Context Protocol , from Anthropic.

The six types of mind maps used in this MCP-Server are:

• presents - Generates a mind map for presentations on a topic;

• compare - Generates a mind map comparing two topics;

• initial - Generates a mental map of initial knowledge about the topic;

• intermediate - Generates a mind map of intermediate knowledge on the topic;

• problems - Generates a mental map of analysis of problems related to the topic;

• review - Generates a mind map to review content on a topic.

Technologies used

Requirements

• Python installed (version 3.10 or higher);

• uv package installed;

• Claude Desktop installed.

How to install on Claude Desktop

Now I will detail how my step by step was in Windows 11, using the terminal (shortcut CTRL + SHIFT + ' ) in VSCode:

1. I have installed the most up-to-date version of Python

2. In VSCode, I used the terminal to check the python version with the command

   python --version

3. So I installed the uv with the remote control

   pip install uv

4. To check if everything was ok, I used the command

   uv

5. To create the project folder, I used this command

   mkdir “C:\Users\meu_usuario\OneDrive\area_de_trabalho\mapas_mentais”

[!IMPORTANT] This does not necessarily mean that you will use the same path, you may want to use another path, like the one below.

  mkdir "C:\Users\seu_usuario\mapas_mentais"

Or you can simply download the zip of this project to your machine via Code > Download ZIP right here on GitHub

6. I named the folder I had just created

   cd “C:\Users\meu_usuario\OneDrive\area_de_trabalho\mapas_mentais”

7. I used the command below to open another VSCode window and continue with the other commands directly in the folder

   code .

[!IMPORTANT] If you don't want to create the folder via terminal, you can create a new folder on your desktop or another location that you easily remember, in order to use the shortcut in VSCode CTRL + O Then just look for the folder you just created, click on it and open it in VSCode. Or just import the complete folder of this repository into your VSCode.

8. Back in the terminal, I used the command below to initialize a new Python project, creating configuration files and dependencies automatically

   uv init

9. I then used the command below to create an isolated Python virtual environment to install project dependencies.

   uv venv

10. To activate .venv, I used the command below

.venv\Scripts\Activate.ps1

11. I added the MCP dependency, which is necessary for the project

uv add mcp[cli]

12. I checked if everything was ok, with the command below

uv run mcp

[!IMPORTANT] If the information below appears on your terminal, everything is fine.

13. To create the server.py file, I used this command

uv init --script server.py

[!TIP] Since you may have already downloaded the folder for this repository, the server.py file will already be there in your VSCode at this point.

14. I installed the json below from MCP-Server directly into the claude_desktop_config.json file

"mapas_mentais": {
  "command": "uv",
  "args": [
    "--directory",
    "C://Users//meu_usuario//OneDrive//area_de_trabalho//mapas_mentais",
    "run",
    "server.py"
  ]
}

[!IMPORTANT] If you have already installed Claude Desktop correctly, follow the path to access the claude_desktop_config.json file on your computer
14th. With Claude Desktop open, use the shortcut CTRL + ,
14b. Click on the Desenvolvedor tab and then click on Editar configuração
14c. Find the claude_desktop_config.json file and edit it in VSCode correctly
14d. Save the file with CTRL + S
14e. Close Claude Desktop and reopen it after a few seconds
14f. Check the configuration icon to see if the MCP "mental_maps" tools are installed correctly

The tools were named `"present", "compare", "initial", "intermediate", "problems" and "review".

Useful links

• Official documentation of the Model Context Protocol - You will know all the details of this innovation from Anthropic

• Anthropic's official website - To stay up to date with the latest news and studies on Claude models

• How to Download Claude Desktop - Direct Download Link

• How to Install VSCode - Direct Download Link

• Official uv package documentation - You will know all the details about uv and how it is important in python

• venv — Creating virtual environments - Full explanation of how venvs work

• AI/LMM Models Icon Set - Very good site to get AI ecosystem icons

• Devicon - very complete site also with general icons about technology

Contributions

Contributions are welcome! If you have ideas to improve this project, feel free to fork the repository.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

Mario Lucio - Deadline®