Pizza API and MCP server for building AI Agents
This Azure sample demonstrates building an AI agent integration end to end: a serverless pizza-ordering API, an MCP server exposing that API to agents such as GitHub Copilot and Claude, and a live dashboard, all runnable locally without any cloud cost.
The sample is a small microservices architecture: a serverless Pizza API for menu, orders and business logic, a Pizza MCP server built on Azure Container Apps that exposes that API's operations as MCP tools such as browsing the menu, placing an order and checking order status, a static-web-app dashboard showing orders in real time, and a separate registration service and app for user sign-up.
Because the MCP server speaks the standard Model Context Protocol, it works with any MCP-compatible client, including GitHub Copilot and Claude, and the whole system can be run and tested locally before anything is deployed to Azure.
It is written in TypeScript and released under the MIT license.