MCP – Model Context Protocol, introduced by Anthropic in 2024, is a standard model for how applications and services can be discovered and interacted with in a universal way. MCP leverages language models to summarize the rich output of these services and can present information in a human-readable form. It is described as being analogous to USB-C, which is a standard for connecting devices and operating systems, connecting agents to software applications and services. Until now, every agentic developer has had to write their own custom plumbing, glue code, etc., to do this. Without a standard like MCP, building real-world agentic workflows is essentially impossible.
MCP provides secure, bi-directional API-level communication between AI large language models (LLMs) and agents, and external data sources or tools. MCP Servers handle requests from AI clients, process them, and return contextual responses. The Anthropic-sourced protocol has added open source contributions and SDKs produced in Python, TypeScript, C#, and Java. There is a GitHub MCP Registry for discovering MCP servers.
