MCP (Model Context Protocol)
Anthropic's open standard for exposing tools, resources, and prompts to AI models — released in late 2024, broadly adopted across the agent ecosystem, the connective tissue of modern AI tool integration.
Production AI is not a prompt. It is a system of context, tools, permissions, traces, evals, and feedback loops.
What it is
The Model Context Protocol (MCP) is an open standard, introduced by Anthropic in November 2024, for how AI models discover and call tools, read resources, and use prompt templates. An MCP server describes its capabilities; an MCP client (a desktop AI app, an IDE, an agent runtime) reads the description and uses the capabilities. It's the closest thing the AI agent era has to a settled interoperability standard.
Why it matters
Before MCP, every integration between an AI app and a tool was custom: the desktop app knew how to call GitHub, the agent runtime knew how to call Slack, the IDE knew how to call Linear, and none of those integrations were portable. MCP makes the integration portable — write one MCP server for your tool and every MCP-aware client can use it.
How it works
MCP servers expose tools (callable functions with JSON Schema), resources (read-only content with URIs), and prompts (templated instructions). Communication is JSON-RPC over stdio or HTTP. The client handles discovery, authentication scope, and the call. The MCP Tool Registry adds governance — which clients can see which servers, with which scope, audited.
Related resources
A governed catalog of every tool an AI agent can call — your APIs, your databases, your internal systems — with typed schemas, permission scopes, audit trails, and the standard protocol (MCP) that turns 'we exposed it to the LLM' into 'we know exactly who called what when'.
The set of open standards that let AI agents talk to tools, to data, and to each other — MCP for tools and resources, A2A for agent-to-agent collaboration, plus the supporting protocols around them.
The mechanism by which a language model invokes external functions — APIs, databases, code execution, retrieval — and reads the results back to continue its work.
A protocol for agents to discover, describe, and call each other directly — Google's open spec for letting independent agent systems collaborate without each one becoming a tool for the other.