Model Context Protocol (MCP)
The Model Context Protocol (MCP) is an open standard for connecting AI agents to external tools and data through a common interface, so any compliant client can use any compliant server.
Before MCP, every agent wired up each tool or data source with bespoke glue code. MCP standardises that connection: a server exposes tools, resources, and prompts over a defined protocol, and any MCP client — an agent or IDE — can discover and call them without custom integration. Introduced by Anthropic in late 2024, it lets one tool server work across many different agents, much as a common port lets any device plug in.
Key points
- An open standard: any MCP client can use any MCP server’s tools, resources, and prompts.
- Replaces bespoke per-tool integration with one shared protocol.
- A server exposes capabilities; the agent discovers and calls them at runtime.
- Introduced by Anthropic (2024) and adopted across a growing tool ecosystem.
Learn it by building
Way of the Agent teaches AI agent engineering level by level — concepts like this one, then hands-on practice, XP, and the AI Chronicles.
Start learning free →