MCP (Model Context Protocol)
An open standard by Anthropic that lets AI assistants connect to external tools, data sources, and services through a unified protocol.
What Is MCP (Model Context Protocol)?
The Model Context Protocol (MCP) is an open-source standard introduced by Anthropic in late 2024 that defines how AI assistants communicate with external systems. Before MCP, every tool integration required bespoke code — a different plugin format for each LLM. MCP standardises this with a client-server architecture: an MCP server exposes capabilities (tools, resources, prompts), and an MCP client (such as Claude Desktop) connects to it.
MCP servers can expose three primitive types: Tools (callable functions like "search the web"), Resources (readable data like files or database rows), and Prompts (reusable prompt templates). This means a single MCP server for GitHub can be used by Claude, any MCP-compatible IDE, or a custom agent without modification.
Why It Matters
MCP is the emerging "USB-C for AI" — a universal connector that lets developers build one integration that works across all MCP-compatible AI systems. It accelerates the agentic AI ecosystem by removing integration lock-in and enabling rich, context-aware AI workflows over real data.
Real-World Examples
Claude Desktop + File System
Anthropic ships an MCP server for local file access. Connect it to Claude Desktop and the assistant can read, write, and navigate your filesystem.