Overview
A comprehensive guide to learning, building, and implementing Model Context Protocol in modern AI applications
Model Context Protocol (MCP)
Welcome to the comprehensive documentation for Model Context Protocol (MCP). Whether you're new to AI development or an experienced programmer, this guide will help you understand and implement MCP in your applications.
What is MCP?
MCP allows AI applications like Claude or GitHub Copilot to securely connect with your data sources and tools. Instead of building custom connectors for every data source, MCP provides a universal way for AI to access different systems while maintaining security and control.
Documentation Roadmap
Introduction
Learn the basics of MCP and why it matters for AI development.
Architecture
Understand how MCP connects clients, servers, and LLMs.
Protocol
Explore the technical details of how MCP handles communications.
Quick Start Guide
Build your first MCP server with step-by-step instructions.
For Different Users
Getting Started with MCP as a User
If you want to use MCP with existing applications like Claude Desktop:
- Download an MCP-compatible application (like Claude Desktop)
- Configure the application to use MCP servers
- Start using AI with access to your data and tools
Check out our section on popular MCP clients to find applications that support MCP.
Getting Started
Choose your path based on your experience level and needs:
New to MCP?
Start with our Introduction to understand the core concepts.
Ready to Explore?
Check out the Architecture to understand how MCP works.
Want to Build?
Jump into our Quick Start Guide to create your first MCP server.
Need Examples?
Browse our collection of example servers and client applications.
MCP is an open standard that's rapidly evolving. This documentation is based on the April 2025 specification. Check modelcontextprotocol.io for the latest updates.
How is this guide?