Model Context Protocol (MCP) Labs

This repository contains a series of labs for learning how to build and use Model Context Protocol (MCP) servers and integrate them with AI Agents.

What is MCP?

The Model Context Protocol (MCP) is a standardized protocol that enables AI models to interact with external tools and data sources. MCP follows a client-server architecture:

• Host: The application that needs AI capabilities
• Client: Part of the host that manages connections to MCP servers
• Server: Provides tools and resources that the AI can use

MCP enables AI models to:

• Execute Tools: Perform actions like searching, calculating, or accessing external systems
• Access Resources: Retrieve data from structured sources via URI templates
• Get Contextual Information: Receive additional context to improve responses

This standardized approach allows AI capabilities to be portable across different platforms and models, creating a consistent interface for AI-powered functionality.

Lab Structure

• Lab 01: Hello Claude - A minimal MCP server with Claude Desktop integration for interactive testing
• Lab 02: Retail MCP Servers - Multiple MCP servers working together for a retail use case
• Lab 03: AWS Cloud Deployment - Deploy MCP servers to AWS Fargate with HTTPS and streaming support
• (More labs will be added in the future)

Getting Started

Each lab directory contains its own README with specific instructions:

1. Start with Lab 01 to learn the basics of MCP server implementation and Claude Desktop integration:

cd lab01-hello-claude

2. Continue with Lab 02 to explore how multiple MCP servers can work together:

cd lab02-retail-mcp-servers

3. Advance to Lab 03 to deploy MCP servers to AWS with Fargate:

cd lab03-aws-cloud-deployment

Be sure to read the README in each lab directory

cat README.md

Resources

• Model Context Protocol Specification
• MCP TypeScript SDK
• Anthropic Claude Documentation