Model Context Protocol (MCP)
An open standard for connecting AI assistants to tools and data.
The Model Context Protocol (MCP) is an open standard that lets AI applications connect to external tools, data and services in a consistent way. Anthropic introduced it on 25 November 2024 as a standard for connecting AI assistants to the systems where data lives.
An MCP server exposes tools and data; an MCP client, such as an AI assistant or coding agent, connects to it and can then use those tools. On 9 December 2025 Anthropic donated MCP to the Agentic AI Foundation, a fund under the Linux Foundation co-founded with Block and OpenAI.
Key characteristics
- One protocol for connecting AI assistants to many tools and data sources.
- Servers expose tools and data; clients such as assistants call them.
- Introduced by Anthropic in November 2024, now under the Linux Foundation's Agentic AI Foundation.
- Supported by many assistants, coding tools and website builders.
Example
A SaaS product ships an MCP server, so a customer can ask their AI assistant to "create an invoice for Acme" and the assistant calls the product's invoice tool directly.
Frequently asked questions
What is an MCP server?
A program that exposes tools and data to AI assistants over the Model Context Protocol, so an assistant can look things up or take actions in your product.
Who owns MCP?
Anthropic created it and, in December 2025, donated it to the Agentic AI Foundation under the Linux Foundation, which it co-founded with Block and OpenAI.