CookHero
CookHero是一个基于 LLM + RAG + Agent + 多模态的智能饮食与烹饪管理平台，支持智能菜谱查询、个性化饮食计划、AI 饮食记录、营养分析、Web 搜索增强，以及可扩展的 ReAct Agent / Subagent 工具体系，帮助厨房新手轻松成为“烹饪英雄”。
What is CookHero?
CookHero is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to cookhero是一个基于 llm + rag + agent + 多模态的智能饮食与烹饪管理平台，支持智能菜谱查询、个性化饮食计划、ai 饮食记录、营养分析、web 搜索增强，以及可扩展的 react agent / subagent 工具体系，帮助厨房新手轻松成为“烹饪英雄”。
CookHero是一个基于 LLM + RAG + Agent + 多模态的智能饮食与烹饪管理平台，支持智能菜谱查询、个性化饮食计划、AI 饮食记录、营养分析、Web 搜索增强，以及可扩展的 ReAct Agent / Subagent 工具体系，帮助厨房新手轻松成为“烹饪英雄”。
This server falls under the Business Applications category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- CookHero是一个基于 LLM + RAG + Agent + 多模态的智能饮食与烹饪管理平台，支持智能菜谱查询、个
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx cookheroConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use CookHero
CookHero is an intelligent cooking and dietary management platform built on LLM, RAG, and multimodal AI. It supports smart recipe search powered by a hybrid vector + BM25 retrieval system over a Milvus vector database, personalized weekly meal planning, AI-driven nutritional analysis from text or uploaded food photos, and web search augmentation via Tavily. The platform exposes a ReAct agent framework with subagent tooling and MCP protocol support, making it extensible and connectable to external AI assistants.
Prerequisites
- Docker and Docker Compose installed (recommended deployment path)
- Python 3.10+ and uv or pip if running without Docker
- API keys for an LLM provider (LLM_API_KEY), vision API (VISION_API_KEY), and Tavily web search (WEB_SEARCH_API_KEY)
- PostgreSQL and Redis (provided automatically via Docker Compose)
- Milvus vector database (provided automatically via Docker Compose)
Clone the repository
Clone the CookHero repository and navigate into the project directory.
git clone https://github.com/Decade-qiu/CookHero.git
cd CookHeroConfigure environment variables
Copy the example environment file and fill in all required API keys and secrets. At minimum you need LLM_API_KEY, FAST_LLM_API_KEY, VISION_API_KEY, WEB_SEARCH_API_KEY, DATABASE_PASSWORD, and JWT_SECRET_KEY.
cp .env.example .env
# Edit .env with your credentials:
# LLM_API_KEY=<your-llm-api-key>
# FAST_LLM_API_KEY=<your-fast-llm-key>
# VISION_API_KEY=<your-vision-api-key>
# WEB_SEARCH_API_KEY=<your-tavily-key>
# DATABASE_PASSWORD=<postgres-password>
# JWT_SECRET_KEY=<random-secret>Start all services with Docker Compose
Launch PostgreSQL, Redis, Milvus, and the CookHero backend and frontend containers in one command from the deployments directory.
cd deployments && docker-compose up -dLoad the recipe knowledge base
Run the recipe data loader script to index recipe content into the Milvus vector store. This step is required for RAG-based recipe search to work.
# If running without Docker:
python -m scripts.howtocook_loaderAccess the web interface
Open a browser and navigate to http://localhost:3000 to access the CookHero frontend. The backend API is available at http://localhost:8000. Register an account and start asking cooking questions.
CookHero Examples
Client configuration
CookHero runs as a self-hosted web application with an MCP-compatible API endpoint. Connect an MCP client to the running backend to enable AI assistant integration.
{
"mcpServers": {
"cookhero": {
"command": "python",
"args": ["-m", "app.mcp_server"],
"env": {
"LLM_API_KEY": "your-llm-api-key",
"FAST_LLM_API_KEY": "your-fast-llm-key",
"VISION_API_KEY": "your-vision-api-key",
"WEB_SEARCH_API_KEY": "your-tavily-key",
"DATABASE_PASSWORD": "your-db-password",
"JWT_SECRET_KEY": "your-jwt-secret"
}
}
}
}Prompts to try
After CookHero is running, use these prompts via the web UI or connected AI assistant.
- "Find me a quick dinner recipe using chicken and broccoli."
- "Create a weekly meal plan targeting 2000 calories per day."
- "Analyze the nutritional content of this meal: rice, grilled salmon, and steamed vegetables."
- "Search the web for the latest trending healthy breakfast recipes."
- "What are the macros in the recipe I just marked as eaten?"Troubleshooting CookHero
Docker Compose fails to start with 'port already in use' errors
Check that ports 5432 (PostgreSQL), 6379 (Redis), 19530 (Milvus), 8000 (backend), and 3000 (frontend) are free. Stop any conflicting services or change the port mappings in docker-compose.yml.
Recipe search returns no results after startup
The Milvus vector index must be populated before search works. Run 'python -m scripts.howtocook_loader' to index the recipe knowledge base. If running in Docker, exec into the backend container first: docker exec -it cookhero-backend bash.
Image-based nutritional analysis fails
Ensure VISION_API_KEY is set to a valid key for your vision model provider (OpenAI Vision API or compatible). Images must be under 10 MB each and the request can include a maximum of 4 images per analysis.
Frequently Asked Questions about CookHero
What is CookHero?
CookHero is a Model Context Protocol (MCP) server that cookhero是一个基于 llm + rag + agent + 多模态的智能饮食与烹饪管理平台，支持智能菜谱查询、个性化饮食计划、ai 饮食记录、营养分析、web 搜索增强，以及可扩展的 react agent / subagent 工具体系，帮助厨房新手轻松成为“烹饪英雄”。 It connects AI assistants to external tools and data sources through a standardized interface.
How do I install CookHero?
Follow the installation instructions on the CookHero GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with CookHero?
CookHero works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is CookHero free to use?
Yes, CookHero is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
CookHero Alternatives — Similar Business Applications Servers
Looking for alternatives to CookHero? Here are other popular business applications servers you can use with Claude, Cursor, and VS Code.
n8n
★ 189.1kA comprehensive MCP server that provides full control over n8n automation workflows through natural language. It offers 43 tools for managing workflows, executions, credentials, and data tables, with safety features like write-mode protection and dou
LobeHub
★ 77.5k🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
Jeecgboot
★ 46.4kAI 低代码平台，「低代码 + 零代码」双模式驱动：低代码一键生成前后端代码，零代码 5 分钟搭建系统，AI Skills 一句话画流程、设计表单、生成整套系统。内置 AI聊天、知识库、流程编排、MCP插件等，兼容主流大模型。引领「AI 生成 → 在线配置 → 代码生成 → 手工合并->AI修改」开发模式，消除 Java 项目 80% 的重复工作，提效而不失灵活。
CowAgent
★ 44.7kCowAgent (chatgpt-on-wechat) 是基于大模型的超级AI助理，能主动思考和任务规划、访问操作系统和外部资源、创造和执行Skills、通过长期记忆和知识库不断成长，比OpenClaw更轻量和便捷。同时支持微信、飞书、钉钉、企微、QQ、公众号、网页等接入，可选择DeepSeek/OpenAI/Claude/Gemini/ MiniMax/Qwen/GLM/LinkAI，能处理文本、语音、图片和文件，可快速搭建个人AI助理和企业数字员工。
Minds Platform
★ 39.2kPlatform dedicated to building an open foundation for applied Artificial Intelligence, designed for people seeking production-ready AI systems they can truly control, extend and deploy anywhere.
Astrbot
★ 32.8kAI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨
Browse More Business Applications MCP Servers
Explore all business applications servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up CookHero in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use CookHero?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.