---
title: "MCP (Model Context Protocol)"
slug: "mcp"
category: "glossary"
tags: ["glossary", "ai-agents", "mcp", "tools", "protocol", "integrations"]
status: "stable"
last_updated: 2026-08-14
summary: "MCP (Model Context Protocol) is Anthropic's open protocol standardizing how AI agents discover and invoke external tools, resources, and prompts over JSON-RPC."
related:
  [
    "[[ai-agents/mcp-servers]]",
    "[[glossary/tool-use]]",
    "[[glossary/function-calling]]",
    "[[ai-agents/multi-agent]]",
    "[[ai-agents/claude-code]]",
    "[[glossary/system-prompt]]",
  ]
---

## Overview

This page is the atomic definition. The MCP deep-dive lives at [[ai-agents/mcp-servers]].

## Definition

Model Context Protocol (MCP) is an open standard introduced by Anthropic in late 2024 that defines a transport-agnostic JSON-RPC interface between AI models (clients) and external capability providers (servers). An MCP server exposes three primitives: Tools (functions the model can invoke), Resources (data the model can read), and Prompts (reusable prompt templates). The model discovers available tools via a `tools/list` call, then invokes them via `tools/call`. The spec defines two official transports: stdio (local subprocess) and Streamable HTTP (remote, JSON-RPC over a single HTTP endpoint with optional SSE streaming). The older standalone HTTP+SSE transport was deprecated when Streamable HTTP shipped in the 2025-03-26 spec revision. MCP decouples tool implementation from model implementation: a single MCP server works with any client that speaks the protocol. Claude Code, Cursor, and several other AI editors ship as MCP clients. The protocol addresses the N*M integration problem by creating a single standard instead of per-model, per-tool adapters.

## When it applies

Use MCP when building tools for AI agents that should be reusable across different clients or models. For single-model integrations where full portability is not needed, [[glossary/function-calling]] via the provider's native API is simpler.

## Example

An MCP server wraps a Postgres database. Claude Code (MCP client) calls `tools/list`, discovers `run_sql`, and uses it to query the schema. The same server works with any MCP-compatible client.

## Related concepts

- [[ai-agents/mcp-servers]] - the full MCP setup guide including server and client configuration.
- [[glossary/tool-use]] - MCP tool calls are a standardized form of tool use.
- [[glossary/function-calling]] - the provider-specific API that MCP generalizes.
- [[ai-agents/multi-agent]] - MCP enables agent-to-agent tool discovery.
- [[ai-agents/claude-code]] - Claude Code is the reference MCP client.

## Citing this term

> See [[glossary/mcp|MCP (Model Context Protocol)]] (llmbestpractices.com/glossary/mcp).

## Related

- [[ai-agents/mcp-servers]]
- [[glossary/tool-use]]
- [[glossary/function-calling]]
- [[ai-agents/multi-agent]]
- [[ai-agents/claude-code]]