coles-woolworths

This MCP server exposes product search capabilities for Coles and Woolworths. It provides tools to query product information and pricing from both retailers in a unified format, helping AI assistants compare availability and prices across the two stores. You can interact with the server through MCP-compatible clients (e.g., Claude Desktop, Cursor) by invoking the available tools and passing in search queries like product names or SKUs. The server is designed to return structured results that include product name, price, store location, and brief details, enabling straightforward integration into chat workflows or decision-support prompts.

Prerequisites:

• Python 3.8 or higher
• uv package manager (installed via the uv bootstrap script)

Install and run locally:

1. Install uv (if not already installed):

   • macOS/Linux: curl -LsSf https://astral.sh/uv/install.sh | sh
   • Windows: powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

2. Create and activate a project directory, then install Python dependencies: uv pip install fastmcp requests python-dotenv

3. Ensure you have a Coles API key and set it in a .env file or as an environment variable: COLES_API_KEY=your_coles_api_key

4. Run the MCP server: uv run main.py

Notes:

• The server reads configuration from environment variables and command-line arguments as shown above.
• You can customize the environment by adding additional variables as needed for your deployment.

Tips and common issues:

• Ensure COLES_API_KEY is valid; without it Coles searches may fail or return limited data.
• When running through Claude Desktop or Cursor, ensure the mcpServers configuration points to the correct path to main.py and that Python/uv is installed on the host.
• If you encounter transport issues, verify that the MCP client is configured to use stdio transport by default unless you’re explicitly enabling socket or HTTP transports.
• For production deployments, consider mounting a virtual environment and pinning dependency versions to avoid breaking changes when dependencies update.
• You can extend tooling by adding more search options (e.g., category, dietary preferences) if supported by the underlying APIs.