Glossary · Agentic AI
What is MCP? Definition and meaning
An open protocol, created by Anthropic, that lets a model use tools, data sources and prompts exposed by servers.
Definition
MCP: An open protocol, created by Anthropic, that lets a model use tools, data sources and prompts exposed by servers.
In short: Model Context Protocol, the open standard for connecting models to tools and data.
When it is used
Connecting agents to databases, APIs and applications through one standard interface.
Where it fits
MCP belongs to Protocols in the Agentic AI topic. These are the other terms in the same group:
- A2A: An open protocol, now under the Linux Foundation, with which agents discover each other and delegate tasks.
- Function calling: A model returns a structured call to a named function with arguments, and the application runs it and returns the result.
- Structured outputs: Constrains generation to a JSON schema or type so that results are always parseable.
- OAuth for tool access: Uses OAuth flows so an agent acts on a user's behalf with scoped, revocable permissions.
Learn more
The full guide to MCP explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.
Previous term: Mastra · Next term: Medallion architecture