使用说明
📡🧠 Gradio MCP AI聊天笔记本
一个Jupyter笔记本就绪单元格，结合了 人工智能模型（Ollama） 随着 实时工具/数据 通过Gradio MCP服务器。
与人工智能聊天，让它自动调用工具，并获得智能答案——所有这些都来自一个笔记本电脑。
受以下MCP CLient的启发：https://medium.com/@jonigl/build-an-mcp-client in minutes-local-ai-agents-just-got-real a10e186a560f
✨ 特性
- 单细胞交互回路：键入问题，获取答案，键入
quit停止 - 工具调用支持：人工智能可以决定运行真实的工具（例如天气、数据库）
- 清洁输出：带格式的答案
rich - 重试并优雅退出：对连接问题具有鲁棒性
- 易于解释和演示：适用于Jupyter/VSCode笔记本
🚀 快速启动
🐍 请确保您已安装Python 3.8+。
1.️⃣ 安装依赖项
pip install ollama mcp rich nest_asyncio
2️⃣ Ensure servers are running
Ollama server (with your chosen model downloaded, e.g., llama3.1:8b)
Gradio MCP server, e.g., http://localhost:7860/gradio_api/mcp/sse
3️⃣ Open the notebook
Copy the big code cell provided.
4️⃣ Run it & chat
Initially prompted: What would you like to know? (or type 'quit' to exit):
Type questions → get answers!
Type quit to stop.
📦 Requirements
Python 3.8+
ollama Python client
mcp Python client
rich
nest_asyncio
🧰 Why use this?
Combines static AI knowledge with live data
Great demo for non-developers and teams
Clean, notebook-friendly workflow
Easy base for building custom AI agents
📝 License
This project is licensed under the MIT License.
🤝 Contributing
Contributions, feedback, and issues welcome!
If you extend the notebook (more tools, richer UI, etc.), feel free to open a PR.
⭐️ If you find this useful, please star the repo!
