snippy

Features • Lab Tutorial • Getting Started • Guidance

Features

• MCP Tool Integration – expose Azure Functions as discoverable MCP tools for AI assistants
• Durable Agents with Microsoft Agent Framework – build stateful AI agents using ChatAgent with automatic conversation history management
• Multi-Agent Orchestration – coordinate specialized agents (DeepWiki, CodeStyle) using Durable Task Scheduler with fan-out/fan-in patterns
• Vector Search on Cosmos DB DiskANN – semantic code retrieval using Azure OpenAI embeddings and low-latency vector indexing
• Monitoring & Observability – track orchestrations in real-time using DTS dashboard (localhost:8082 local, Azure portal for cloud)
• One-click Deploy – azd up provisions and deploys complete infrastructure including Functions, Cosmos DB, Azure OpenAI, and DTS
• Codespaces & Dev Containers – fully configured development environment in your browser or local VS Code

Tool Matrix

| Tool Name | Purpose | | -------------------------------- | ------------------------------------------------------------------- | | save_snippet | Save code snippets with vector embeddings for semantic search | | get_snippet | Retrieve previously saved code snippets by their unique name | | code_style | Generate language-specific code style guides from saved snippets | | deep_wiki | Create comprehensive wiki documentation by analyzing code snippets | | generate_comprehensive_documentation | Orchestrate multi-agent workflow to produce deep wiki and style guide |

Lab Tutorial

New to Snippy? Start with our comprehensive hands-on lab tutorial that guides you through building the entire application from scratch:

📚 Snippy Tutorial: Building an AI-Enhanced Code Snippet Manager

The tutorial covers:

• Step-by-step implementation – Build the complete application with detailed explanations
• Code review exercises – Understand MCP tools, durable agents, and vector search patterns
• Local development – Set up and test with Docker-based emulators
• Azure deployment – Provision and deploy to production with azd
• Multi-agent orchestration – Monitor complex workflows with DTS dashboard
• End-to-end testing – Test with REST Client, GitHub Copilot, and MCP tools

Additional Resources:

• Quick Reference – Essential commands and common tasks
• Troubleshooting Guide – Solutions to common issues

Whether you're new to Azure Functions, MCP tools, or AI agent development, the lab provides a structured learning path with hands-on experience.

Getting Started

Prerequisites

• Azure subscription with permissions to create resources
• azd CLI installed
• Choose your development environment:
  • GitHub Codespaces (fastest, no local setup required)
  • VS Code Dev Containers (requires Docker Desktop)
  • Local development (requires Python 3.11, Node 18+, Azure Functions Core Tools v4)

Deploy to Azure

Using GitHub Codespaces:

Click the "Open in GitHub Codespaces" badge above, then run:

azd auth login --use-device-code
azd up

Using Dev Containers or Local Environment:

azd init --template Azure-Samples/snippy
azd auth login
azd up

The azd up command will:

• Prompt for an Azure region (recommend eastus or swedencentral)
• Provision all resources (Functions, Cosmos DB, Azure OpenAI, DTS)
• Create an Azure AD app registration for authentication
• Deploy the application code

When complete, the Function App URL and MCP endpoint will be displayed. See the tutorial for detailed instructions on local development, testing with emulators, and monitoring orchestrations.

Guidance

Region Availability

Azure OpenAI model support varies by region. Verify availability here and choose the same region for all Azure resources. eastus and swedencentral are good default choices.

Security

Snippy uses User-Assigned Managed Identity for secure service-to-service authentication. The infrastructure is configured with:

• User-Assigned Managed Identity on the Function App with appropriate RBAC roles:
  • Cosmos DB Data Contributor
  • Storage Blob Data Owner and Queue Data Contributor
  • Application Insights Monitoring Metrics Publisher
  • Azure AI Project Developer

For production deployments, we recommend:

• Restrict inbound traffic with Private Endpoints + VNet integration
• Enable network security features like service endpoints and firewall rules

Contributing

Standard fork → branch → PR workflow. Use Conventional Commits (feat:, fix:) in commit messages.

License

MIT © Microsoft Corporation