gen-ai-starter

概述

这是一个全面的生成式AI入门工具包，集成了多个开源工具和模型，包括Ollama、Open WebUI、LangChain、Streamlit、Chroma和PGVector。它使用Docker和Docker Compose对应用进行容器化。该项目特别支持模型上下文协议(MCP)服务器，能够从OpenAPI规范实现标准化工具加载。项目包含从PokéAPI OpenAPI规范生成的完整MCP服务器，展示了如何将REST API转换为标准化MCP工具并独立部署。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• pokemon_mcp

  Query PokéAPI to get information about Pokémon

  试试:Tell me about Pikachu试试:What are Electric type weaknesses?
• ollama_list_models

  List available Ollama models

  试试:What models are available in Ollama?试试:Show me all installed models
• create_github_issue

  Create a new GitHub issue

  试试:Create a new issue about documentation试试:Report a bug in the application
• search_brave

  Search the web using Brave Search API

  试试:Search for recent news about AI试试:Find information about quantum computing
• explain_code

  Explain a specific file or code pattern

  试试:Explain src/example/agentic_chat.py试试:How does the RAG pattern work?
• guide_learning

  Get structured guidance through learning topics

  试试:Guide me through RAG试试:Teach me agent development basics
• create_learning_path

  Create personalized learning progressions

  试试:Learning path for beginners试试:Path to master RAG techniques
• implement_feature

  Get help implementing new features

  试试:How to add a new vector database example试试:Implement a custom retriever

说明：Tools were inferred from documentation examples and usage patterns rather than explicit MCP tool listings. The actual MCP tools are likely generated from OpenAPI specifications as mentioned in the README.

可对比工具

安装

FAQ

该项目中的MCP服务器有什么用途？

MCP服务器能够从OpenAPI规范实现标准化工具加载，使开发人员可以将REST API转换为MCP工具，供AI代理使用。

如何为我的 own API 创建MCP服务器？

您可以使用openapi-mcp-generator CLI工具。通过'npm install -g openapi-mcp-generator'安装，然后使用'openapi-mcp-generator --input your-openapi.json --output your-mcp-server'生成您的服务器。

gen-ai-starter 对比