Glossary

Model Context Protocol (MCP)

Model Context Protocol (MCP) is an open standard, introduced by Anthropic in November 2024, that defines how AI applications connect to outside tools and data through MCP servers, so one integration works across any compatible AI client.

What Is Model Context Protocol (MCP)?

In short MCP uses a client-server design.

An MCP server wraps a system, such as a CRM, a database or a file store, and exposes its data, tools and reusable prompts. An MCP client inside an AI application, such as Claude, ChatGPT, Visual Studio Code or Cursor, discovers what the server offers and calls it, with the server running on your machine or remotely over HTTP.

Before MCP, each AI app needed its own connector for each tool. MCP turns that into one server per tool that works with every compatible client; the MCP site compares it to a USB-C port for AI applications. Anthropic introduced and open-sourced it on November 25, 2024. On December 9, 2025, Anthropic donated MCP to the Agentic AI Foundation, a directed fund under the Linux Foundation co-founded by Anthropic, Block and OpenAI.

Model Context Protocol Example

For example, a founder connects a CRM's MCP server and an outbound platform's MCP server to Claude, then asks which of last week's demo no-shows have not had a follow-up. Claude queries the CRM, finds 7 contacts, drafts a short note for each and, after the founder approves, adds them to a follow-up sequence.

Model Context Protocol in Practice

With MCP, an AI assistant can act in your sales tools, not just talk about them. With MCP servers for the CRM, the outbound platform and a data source, a rep can ask an assistant to find 50 heads of finance at Series B fintech companies, add them to a sequence and log it in the CRM, and the assistant works through each system's own permissions. A server acts with the access you grant it, so connect only servers you trust and check which tools can change data.

FAQ

What is MCP in AI?

Model Context Protocol, an open-source standard for connecting AI applications such as Claude or ChatGPT to external data sources, tools and workflows through MCP servers.

Who created the Model Context Protocol?

Anthropic introduced MCP as an open-source standard on November 25, 2024. In December 2025 it was donated to the Agentic AI Foundation under the Linux Foundation, and Anthropic said its governance model would stay unchanged.

What is the difference between MCP and an API?

An API is one system's interface for other software. MCP is a standard layer that lets any AI client discover and call tools the same way, and many MCP servers wrap an existing API.

Is MCP safe to use?

It is as safe as the servers you connect. A server can read and change data with the access you grant it, so use servers from vendors you trust, prefer OAuth sign-in over shared keys, and review which tools can write or send.

Do This in Ken

Do This in Ken

Add the Ken MCP server to Claude, ChatGPT, Cursor or another MCP client, then work with Ken in plain language from that client.

Module
Ken MCP server
Open in Ken
/mcp

The new Ken opens to the waiting list in batches.

Sources

Every source, with the date we checked it
SourceChecked
Anthropic: Introducing the Model Context Protocol
Anthropic: Donating the Model Context Protocol and establishing the Agentic AI Foundation
modelcontextprotocol.io: What is the Model Context Protocol?

Changelog

  1. Page published.