snippy

Overview

Snippy is a comprehensive MCP server built on Azure Functions that demonstrates how to create AI-powered tools for code snippet management. It integrates Azure OpenAI for embeddings, Cosmos DB with vector indexing for semantic search, and Durable Functions for multi-agent orchestration. The server exposes multiple MCP tools including save_snippet, get_snippet, code_style, and deep_wiki, enabling AI assistants like GitHub Copilot to interact with a sophisticated code management system.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• save_snippet

  Save code snippets with vector embeddings for semantic search

  Try:Save this Python code for a bubble sortTry:Store this JavaScript function for future reference
• get_snippet

  Retrieve previously saved code snippets by their unique name

  Try:Get my bubble sort snippetTry:Retrieve the JavaScript function I saved earlier
• code_style

  Generate language-specific code style guides from saved snippets

  Try:Create a Python style guide from my saved snippetsTry:Generate best practices for JavaScript coding from my snippets
• deep_wiki

  Create comprehensive wiki documentation by analyzing code snippets

  Try:Generate documentation for my REST API snippetsTry:Create a wiki from all my Python function snippets
• generate_comprehensive_documentation

  Orchestrate multi-agent workflow to produce deep wiki and style guide

  Try:Create comprehensive documentation for my entire codebaseTry:Generate both wiki and style guide for my project

Comparable tools

Installation

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

{
  "mcpServers": {
    "snippy": {
      "command": "python",
      "args": ["-m", "snippy.mcp"]
    }
  }
}

FAQ

What is the primary purpose of Snippy?

Snippy is an MCP server that exposes Azure Functions as tools for AI assistants to manage code snippets with semantic search and generate documentation.

Can Snippy be deployed on-premises?

Snippy is designed for Azure deployment but can be developed locally using Docker emulators before deploying to Azure.

Compare snippy with