uml-mcp

Installation

This guide explains how to install and set up the UML-MCP server.

System Requirements

• Python 3.10 or higher
• uv, Poetry, or pip
• Optional: Docker for running local PlantUML or Kroki servers

Installation Steps

1. Clone the repository:

git clone https://github.com/yourusername/uml-mcp.git
cd uml-mcp

1. Install the dependencies:

With uv (recommended):

uv sync

With Poetry:

poetry install

With pip:

pip install -r requirements.txt

1. For development (tests, linting):

uv sync --all-groups
# or: poetry install --with dev
# or: pip install -r requirements-dev.txt

Verifying Installation

To verify your installation:

python server.py --list-tools

You should see output similar to:

Starting UML-MCP Server v1.2.0
Server Name: UML Diagram Generator
Available Tools: 12
Available Prompts: 3

IDE Integration

Cursor and other MCP clients

Configure your IDE to run the UML-MCP server. Use manual configuration:

• Command: python (or full path to your venv Python if using uv/Poetry)
• Arguments: ["/path/to/uml-mcp/server.py"] — use the full path to server.py
• Working directory: /path/to/uml-mcp (project root)
• Output directory: Optional; set MCP_OUTPUT_DIR in env or use default ./output

See config/README.md for where each app stores its config, and Cursor integration or Claude Desktop integration for step-by-step setup.

Optional Components

Local Diagram Servers

For better performance or offline use, you can set up local servers:

PlantUML Server

docker run -d -p 8080:8080 plantuml/plantuml-server

Kroki Server

docker run -d -p 8000:8000 yuzutech/kroki

Troubleshooting

If you encounter issues during installation:

1. Ensure Python 3.10+ is installed and in your PATH
2. Check that all required dependencies are installed
3. Verify any local servers are running correctly
4. Ensure proper permissions for the output directory

This site is open source. Improve this page.