π²
@Aryan-Jhaveri
A MCP Server for Canada's Food Guide - To allow LLM's with preset webscraping tools to fetch recipes.
Overview
What is π²?
An MCP (Model Context Protocol) server that wraps a web scraper built specifically for Canada's Food Guide, exposing recipes and filters as ready-to-use tools for AI assistants.
How to use π²?
Install Python 3.8 or higher, clone the repository, install dependencies with pip3 install -r requirements.txt, then configure Claude Desktop with the absolute paths to your Python executable and src/server.py in the JSON settings.
Key features of π²
- Search recipes by text, ingredients, meal type, appliance, and collections
- Fetch complete recipe details including ingredients, instructions, and tips
- List all available filter categories and their options
- Configure maximum search pages to limit results
- Downloadable recipe output as JSON or Markdown
Use cases of π²
- Find themed recipes (e.g., vegetarian or kid-friendly) for meal planning
- Search for recipes that match ingredients you already have at home
- Retrieve full cooking instructions and nutritional info for a chosen recipe
- Discover available filter values before building a search query
- Automate recipe retrieval for integration with calendar or grocery apps
FAQ from π²
What does MCP stand for?
MCP stands for Model Context Protocol, which allows AI assistants to use external tools like this server.
What are the system requirements?
Python 3.8 or higher and pip (Python package installer). The server requires an internet connection to access Canada's Food Guide.
How does π² differ from a direct API?
It uses web scraping with BeautifulSoup4, so it is slower and depends on the Canada's Food Guide website structure remaining unchanged.
What known limitations exist?
Maximum search pages and results are capped. Data inconsistency may occur if the Food Guide site has editing anomalies (e.g., missing recipes between filtered and unfiltered searches).
How can I test the server without Claude?
You can run main.py as a command-line tool, or use cli.py for a terminal-based search interface for debugging.