Model Context Protocol (MCP)
An open standard that lets AI models connect to external tools and data sources through a unified interface.
What It Is
Model Context Protocol (MCP) is a standard created by Anthropic that defines how AI models communicate with external tools, databases, and services. Before MCP, every tool integration required custom code for each AI model and each service. MCP creates a common language so that any MCP-compatible AI tool can connect to any MCP-compatible service without custom wiring. Think of it like USB for AI: a single standard interface that lets different devices (tools and data sources) plug into different computers (AI models) without needing a unique cable for each combination.
Why It Matters
MCP is what makes modern AI assistants genuinely useful beyond conversation. When Claude Code connects to your file system, reads your GitHub repositories, or queries your Supabase database, it is using MCP servers to do so. For operators, MCP means you can extend your AI tools with new capabilities by adding MCP servers rather than writing custom integrations. The ecosystem of available MCP servers is growing rapidly, covering everything from databases to design tools to communication platforms.
In Practice
When you configure Claude Code to access your file system or connect to Notion, you are setting up MCP servers. Each server exposes specific capabilities (read files, search documents, create records) that the AI can then use as tools during your conversation. Operators can install community-built MCP servers or configure official ones to expand what their AI assistant can do.