Snippy Lab - Build an AI-Powered Code Snippet Manager

Welcome! This hands-on lab teaches you to build Snippy, an intelligent code snippet manager using Azure Functions, Azure OpenAI, and the Model Context Protocol (MCP).

What You'll Build

• Serverless API with Azure Functions for saving/retrieving code snippets
• Vector Search using Azure OpenAI embeddings and Cosmos DB
• MCP Tools that GitHub Copilot can discover and use
• AI Agents that generate documentation and style guides from your code
• Multi-Agent Orchestration using Durable Functions

Prerequisites

Before starting, ensure you have:

Required Software

• Visual Studio Code - Download
• Python 3.11+ - Download
• Git - Download
• Azure CLI - Install
• Azure Functions Core Tools v4 - Install
• Azure Developer CLI (azd) - Install
• Docker Desktop - Download

Recommended

• uv package manager (faster than pip) - Install

Required Accounts

• Azure Subscription - Create free account
• GitHub Account with Copilot - GitHub Copilot

Quick Start

1. Clone this repository

   git clone https://github.com/Azure-Samples/snippy.git
   cd snippy
   

2. Follow the tutorial

3. Open TUTORIAL.md for step-by-step instructions

4. Refer to TROUBLESHOOTING.md if you encounter issues

5. Clean up when done

   azd down --purge --force
   

What You'll Learn

• Azure Functions as an AI integration platform
• Model Context Protocol (MCP) for AI assistant integration
• Vector embeddings and semantic search with RAG patterns
• Microsoft Agent Framework for building AI agents
• Durable Functions for multi-agent orchestration
• Azure OpenAI integration with embeddings and chat models

Support

• Issues: Found a bug? Open an issue
• Questions: Check TROUBLESHOOTING.md first
• Discussions: GitHub Discussions

License

This project is licensed under the MIT License - see the LICENSE.md file for details.