Model Context Protocol (MCP)

Full definition

MCP defines two roles: clients (AI apps like Claude Desktop, IDEs, agent frameworks) and servers (wrappers around tools, databases, APIs, file systems). Servers expose tools, resources, and prompts; clients discover and invoke them over stdio or HTTP+SSE. As of 2025, MCP is supported by Anthropic Claude, OpenAI, Google, and an ecosystem of 1,000+ community servers (GitHub, Slack, Postgres, Notion, etc.). MCP standardizes what previously required bespoke per-app integrations.

Why it matters

MCP turns tool integration from N×M to N+M — every MCP-compliant model can use every MCP server. For enterprises, this dramatically lowers the cost of connecting agents to internal systems and reduces vendor lock-in.

Example

A consulting firm exposes its CRM, file system, and time-tracking tools as three MCP servers; Claude, Cursor, and an internal LangGraph agent all consume the same servers without bespoke integrations.

Related terms

• Function Calling
  Function calling is an LLM capability where the model — given a schema of available functions — emits a structured invocation (function name and JSON arguments) instead of free-form text, letting downstream code execute the call deterministically.
• Tool Router
  A tool router is a layer in an agent system that selects which tool (or subset of tools) to expose to the model for a given query, instead of putting all tool definitions in every prompt.
• Tool Use (LLM)
  LLM tool use (also called function calling) is the ability of a language model to invoke external functions, APIs, databases, or systems based on the user's request — turning the LLM from a text generator into a controller that can act on the world.
• AI Agent / Agentic AI
  An AI agent is an LLM-driven system that, given a goal, plans a sequence of steps, calls tools or APIs, observes the results, and iterates until the goal is reached — going beyond single-turn chat to multi-step autonomous action.