The Model Context Protocol (MCP) is an open standard for connecting AI applications to external data sources, tools and workflows.
MCP was introduced by Anthropic in late 2024 and is now maintained as an open standard at modelcontextprotocol.io. It defines how a client (the AI app) discovers and calls capabilities on a server (your data or tool). Servers expose tools the model can call, resources it can read and prompts it can use. The protocol handles the plumbing so a server written once works with any compliant client.
Support is broad: Claude, ChatGPT, Visual Studio Code and Copilot, Cursor and many others act as MCP clients. For memory and knowledge products this matters because it removes the per-app integration problem. A single local server can answer Claude Desktop, Claude Code and the ChatGPT desktop app from the same files. Capabilities still vary by client, so a server may offer tools that one client uses and another ignores.