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:

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

cd lab01-hello-claude

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

cd lab02-retail-mcp-servers

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

Model Context Protocol (MCP) Labs