mcp-registry

john-carroll-sw/mcp-registry

3.2

If you are the rightful owner of mcp-registry and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

This document provides a comprehensive guide to building, deploying, and integrating a Model Context Protocol (MCP) server using Azure API Center.

MCP Registry Lab: End-to-End Azure API Center MCP Integration

This repository demonstrates a complete hands-on workflow for working with Model Context Protocol (MCP) servers using Azure API Center's MCP registry. You'll build, deploy, register, discover, and consume MCP tools in AI applications.

Overview

This lab covers four main parts:

  1. Build a tiny MCP server with FastMCP - Create a weather MCP server
  2. Containerize and deploy on Azure Container Apps (ACA) - Deploy with consumption plan (scales to 0, generous free grant)
  3. Register the server in Azure API Center's MCP registry - Make it discoverable
  4. Discover and use MCP tools from API Center Portal - Connect with LangGraph and consume tools

Prerequisites

  • Azure subscription with API Center enabled
  • Python 3.10+
  • Azure CLI
  • VS Code with MCP extension (optional)
  • OpenAI or Azure OpenAI access

Part 1: Build and Deploy MCP Server

See for complete instructions on:

  • Creating a weather MCP server with FastMCP
  • Deploying to Azure Container Apps
  • Testing the deployment

The lab shows how to deploy a server that provides weather forecast tools accessible via SSE (Server-Sent Events) with API key authentication.

Part 2: Register in Azure API Center's MCP Registry

Follow the official Microsoft documentation to register your deployed MCP server:

📖 Set up API Center Portal

This step makes your MCP server discoverable through the Azure API Center Portal preview.

Part 3: Discover and Use MCP Tools

See for complete instructions on:

  • Setting up the LangGraph application with MCP integration
  • Configuring Azure OpenAI with Azure AD authentication
  • Running the weather query example
  • Troubleshooting common issues

Architecture

[LangGraph Agent] 
    ↓
[langchain-mcp-adapters]
    ↓  
[MCP Server on ACA] ←→ [Azure API Center Registry]
    ↓
[Weather API/Tools]

Key Features Demonstrated

  • Remote MCP Integration: Connect to MCP servers over HTTPS/SSE
  • Azure AD Authentication: Secure connection to Azure OpenAI using managed identity
  • Tool Discovery: Automatic discovery of MCP tools and their schemas
  • Reactive Agents: LangGraph agents that can reason about and use MCP tools
  • Environment-based Configuration: Clean separation of secrets and config
  • Production Deployment: Container-based deployment on Azure with scaling

Next Steps

  • Explore the Azure API Center Portal preview to discover other MCP servers
  • Add additional MCP servers to your MultiServerMCPClient configuration
  • Implement error handling and logging for production use
  • Experiment with different LangGraph agent patterns
  • Try different Azure OpenAI models and configurations

Resources

This lab provides a complete foundation for building AI applications that leverage distributed MCP tools through Azure's cloud infrastructure and API management capabilities.