What is MCP (Model Context Protocol)?
Open standard that lets AI agents talk to external tools and data sources through a uniform server interface. Built by Anthropic, now supported by OpenClaw, Hermes, Claude Cowork, and IronClaw. Replaces per-tool integrations with one protocol.
What it does
MCP is an open-source standard for connecting AI applications to external systems: data sources such as local files and databases, tools such as search engines and calculators, and workflows such as specialised prompts. The MCP project describes it as a USB-C port for AI applications: one standard connector instead of a custom integration for every pairing of app and service.
Who builds with it
There are two sides. Servers expose data and tools; clients are the AI applications that connect to them. The MCP documentation lists Claude, ChatGPT, Visual Studio Code and Cursor among the products that support it, so a server built once can be used from several of them.
Why it matters for agents
An agent is only as useful as what it can reach. MCP is how most agent platforms attach calendars, notes, repositories, design files and databases without bespoke code. The flip side is that every connected server is also a path for untrusted content and powerful actions to reach the model, so the MCP project publishes its own security best-practices guide; read it before connecting a server you did not write.
Example
A coding agent connects to a Figma server over MCP and can then read a design file and generate the matching web app, instead of someone pasting screenshots into a chat. The same server works from any client that speaks MCP.
Sources
See also
← Back to the full AI agent glossary.