Open Data MCP Server

What is Open Data MCP?

Open Data MCP is a framework that connects open data sources to Large Language Models through the Model Context Protocol (MCP). It enables two primary functions:

1. Access to Open Data: Easily connect your LLM applications to public datasets
2. Publishing Open Data: Contribute and distribute your own datasets to the community

Installation

To use Open Data MCP, you'll need to install the CLI tool and set up the necessary dependencies:

# Install uv package manager (required)
# macOS
brew install uv

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

# Linux/WSL
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install Open Data MCP
pip install odmcp

Using Open Data MCP

Accessing Open Data

The CLI tool provides several commands to work with data providers:

# List all available data providers
uvx odmcp list

# Get information about a specific provider
uvx odmcp info PROVIDER_NAME

# Set up a provider's MCP server for Claude Desktop app
uvx odmcp setup PROVIDER_NAME

# Remove a provider from Claude Desktop app
uvx odmcp remove PROVIDER_NAME

Example: Setting up Swiss Railway (SBB) Data

# Make sure Claude Desktop app is installed
uvx odmcp setup ch_sbb

After setup, restart Claude and you'll see a new hammer icon at the bottom right of the chat interface. You can now ask questions about SBB train network disruptions and other related information.

Contributing Your Own Data Provider

If you want to contribute by building and publishing your own public dataset:

1. Clone the repository:

git clone https://github.com/OpenDataMCP/OpenDataMCP.git
cd OpenDataMCP

2. Set up the development environment:

uv venv
source .venv/bin/activate  # Unix/macOS
# or
.venv\Scripts\activate     # Windows
uv sync
pre-commit install

3. Create a new provider module in src/odmcp/providers/ following the naming pattern {country_code}_{organization}.py

4. Implement your provider using the template file as a guide, defining appropriate Tools and Resources

5. Add tests for your provider in the tests/ directory

6. Test your MCP server using the experimental client: uv run src/odmcp/providers/client.py

Troubleshooting

• If you encounter issues with Claude Desktop app not showing the MCP tools, try restarting the app after setup
• For Windows users, ensure that uv is properly installed and available in your PATH
• Check the Discord community for help with specific providers or implementation questions

Community Support

Join the Discord community at https://discord.gg/hDg4ZExjGs for help, discussions, and collaboration opportunities.