Content
# HowToCook-py-MCP -- A Delicious Week Ahead, No More Worrying About Meals
[English](./README_EN.md) |
> Let your AI assistant become your personal chef and plan your meals!
This is a Python version of the MCP server for recipe assistance, using the FastMCP library. Built on top of [Anduin2017/HowToCook](https://github.com/Anduin2017/HowToCook), this project enables your AI assistant to recommend recipes, plan meals, and solve the eternal question of "what to eat today"!
Special thanks to [worryzyy/HowToCook-mcp](https://github.com/worryzyy/HowToCook-mcp), from which this Python version was inspired 😄.
## 📸 Preview

## 🔌 Supported MCP Clients
This server has been tested with the following clients:
- 📝 Cursor
## ✨ Delicious Features
This MCP server offers the following culinary tools:
1. **📚 Get All Recipes** (`get_all_recipes`) - Returns simplified data for all available recipes -- Use with caution due to large context
2. **🔍 Get Recipes by Category** (`get_recipes_by_category`) - Query recipes by specified category, e.g., seafood, breakfast, meat dishes, or main courses!
3. **🎲 What to Eat** (`what_to_eat`) - A solution for indecisiveness! Recommend today's menu based on the number of people
4. **🧩 Recommend Meal Plans** (`recommend_meals`) - Plan a week's worth of delicious meals based on dietary restrictions, allergies, and the number of people
## 🚀 Quick Start
### 📋 Prerequisites
- Python 3.12.9+ 🐍
- Necessary Python dependencies 📦
### 💻 Installation Steps
1. Clone the repository
```bash
git clone https://github.com/DusKing1/howtocook-py-mcp.git
cd howtocook-py-mcp
```
2. Install dependencies
```bash
pip install -r requirements.txt
```
### ❓ Why Not Use uv?
You forget a thousand things every day; why not forget this one too?
## 🍽️ Getting Started
### 🔥 Start the Server
```bash
# Ensure you're in the project root directory
python -m src.app
```
The server will run on port 9000 using the SSE protocol.
### 🔧 Configure MCP Client
#### Use Cursor for a Quick Experience
Add the MCP server configuration in Cursor settings:
```json
{
"mcpServers": {
"how to cook": {
"url": "http://localhost:9000/sse"
}
}
}
```
#### Other MCP Clients
For other clients supporting the MCP protocol, refer to their documentation for configuration.
## 🧙♂️ Menu Magic Usage Guide
Example prompts for use in various MCP clients:
### 1. 📚 Get All Recipes
No parameters needed; just summon the recipe book!
```
Use the howtocook-py-mcp MCP service to query all recipes
```
### 2. 🔍 Get Recipes by Category
```
Use the howtocook-py-mcp MCP service to query seafood recipes
```
Parameters:
- `category`: Recipe category (seafood, breakfast, meat dishes, main courses, etc.)
### 3. 🎲 What to Eat?
```
Use the howtocook-py-mcp MCP service to recommend a menu for 4 people for dinner
```
Parameters:
- `people_count`: Number of people (1-10)
### 4. 🧩 Recommend Meal Plans
```
Use the howtocook-py-mcp MCP service to recommend a week's recipes for 3 people; we don't eat cilantro and are allergic to shrimp
```
Parameters:
- `allergies`: List of allergens, e.g., ["garlic", "shrimp"]
- `avoid_items`: Ingredients to avoid, e.g., ["onion", "ginger"]
- `people_count`: Number of people (1-10)
## 📝 Tips
- This service is compatible with all AI assistants and applications supporting the MCP protocol
- The AI may take some time to get familiar with using these tools upon first use
## 📄 Data Source
Recipe data comes from a remote JSON file at:
`https://mp-bc8d1f0a-3356-4a4e-8592-f73a3371baa2.cdn.bspapp.com/all_recipes.json`
## 🤝 Contributions
Welcome to fork and submit pull requests to improve this culinary assistant!
## 📄 License
MIT License - Use freely, just like sharing recipes!
---
> 🍴 Deliciousness is about to begin; are your taste buds ready?
Connection Info
You Might Also Like
markitdown
Python tool for converting files and office documents to Markdown.
OpenAI Whisper
OpenAI Whisper MCP Server - 基于本地 Whisper CLI 的离线语音识别与翻译,无需 API Key,支持...
oh-my-opencode
Background agents · Curated agents like oracle, librarians, frontend...
claude-flow
Claude-Flow v2.7.0 is an enterprise AI orchestration platform.
chatbox
User-friendly Desktop Client App for AI Models/LLMs (GPT, Claude, Gemini, Ollama...)
continue
Continue is an open-source project for seamless server management.