---
title: Model Context Protocol (MCP) — AI Glossary
description: The Model Context Protocol is an open standard for connecting AI models to external tools, data sources, and systems through a common interface. It lets any compliant client and server exchange context, tools, and resources, so integrations are reusable across applications rather than rebuilt for each model or vendor.
canonical: https://prompeteer.ai/glossary/model-context-protocol
---

# Model Context Protocol (MCP)

The Model Context Protocol is an open standard for connecting AI models to external tools, data sources, and systems through a common interface. It lets any compliant client and server exchange context, tools, and resources, so integrations are reusable across applications rather than rebuilt for each model or vendor.

## Definition

MCP, introduced by Anthropic and adopted broadly, defines a client-server protocol in which MCP servers expose tools, resources, and prompts, and MCP clients — such as AI applications and agents — discover and call them. This decouples what a model can do from how each capability is implemented.

By standardizing the connection layer, MCP reduces the combinatorial cost of wiring many models to many tools. A single MCP server for a data source can serve any MCP-aware client, which is why it has become a common foundation for agentic and tool-augmented systems.

## Sources

- [Model Context Protocol — Specification](https://modelcontextprotocol.io/specification)
- [Anthropic — Introducing the Model Context Protocol](https://www.anthropic.com/news/model-context-protocol)

## Related terms

- [tool-use](https://prompeteer.ai/glossary/tool-use)
- [function-calling](https://prompeteer.ai/glossary/function-calling)
- [agent](https://prompeteer.ai/glossary/agent)
- [grounding](https://prompeteer.ai/glossary/grounding)