UML

Generates UML diagrams from PlantUML code or natural language via MCP, returning code, URL, and a local image path.
  • python

28

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "swayingleaves-uml-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/Users/yourpath/UML-MCP-Server",
        "run",
        "uml_mcp_server.py"
      ]
    }
  }
}

UML-MCP Server helps you generate UML diagrams from natural language prompts or direct PlantUML code and exposes this capability through the MCP (Model Context Protocol). You can run it locally, integrate it with MCP clients, and save outputs like PlantUML code, a shareable URL, and a local image file path for easy viewing and sharing.

How to use

You can generate UML diagrams by sending PlantUML code directly to UML-MCP Server or by providing descriptive text that the server converts into diagrams. When you request a diagram, you receive three useful results: the PlantUML code, a URL to view the diagram online, and a local file path where the image is saved. If you want to share or view diagrams with others, use the PlantUML URL or copy the PlantUML code into other PlantUML tools.

How to install

# Prerequisites: Python and virtual environment support
# 1) Clone the project
git clone https://github.com/yourusername/UML-MCP-Server.git
cd UML-MCP-Server

# 2) Create and activate a virtual environment
python -m venv uml-mcp-venv
source uml-mcp-venv/bin/activate  # Linux/Mac
# Or
uml-mcp-venv\Scripts\activate  # Windows

# 3) Install dependencies
pip install -r requirements.txt

# 4) Run the MCP server (example command; adjust as needed for your setup)
uv run uml_mcp_server.py

If you plan to integrate with a specific MCP client, ensure you have the appropriate client installed and configured to communicate with this server using MCP-compatible requests.

Additional configurations and notes

Cursor can run the UML-MCP Server locally as a stdio MCP server. Use the following configuration in Cursor to start the server and output diagrams to a dedicated directory.

{
  "mcpServers": {
    "uml_mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/Users/yourpath/UML-MCP-Server",
        "run",
        "uml_mcp_server.py"
      ],
      "output_dir": "/Users/yourpath/uml-output"
    }
  }
}

Troubleshooting

If you encounter issues, check the server logs for error messages, verify that dependencies are installed, ensure the PlantUML endpoint is reachable if you rely on a remote render service, and confirm you have write permissions for the output directory.

Examples and outputs

When you generate a diagram, you typically receive three outputs in response: the PlantUML code, a PlantUML URL to view the diagram, and a local file path where the image is saved. You can view the diagram by opening the URL in a browser or by viewing the image at the local file path.

Available tools

generate_uml

Takes PlantUML code as input and returns the PlantUML code, a shareable PlantUML URL, and a local file path for the generated diagram. Supports saving locally and sharing via URL.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational