What is MCP (Model Context Protocol)?

The Model Context Protocol (MCP) is an open standard for connecting AI models and agents to external tools, data sources and services through a single, uniform interface. Instead of writing a bespoke integration for every tool, a developer runs an MCP server that exposes its capabilities, and any MCP-aware client can use it.

The shape of it

   AI client (IDE, agent, chat)
            │  MCP (JSON-RPC over stdio or HTTP)
   ┌────────┼─────────┐
 server   server    server
(GitHub) (Postgres) (filesystem)

An MCP server advertises three things: tools (functions the model can call), resources (data it can read), and prompts (reusable templates). The client handles discovery, permission and the call loop.

Why it matters

It turns tool integration from an N×M problem into an N+M one, and it makes an agent's capabilities portable across clients. Adoption moved fast through 2025-2026, from IDEs into OS-level agent runtimes.

The security caveat

An MCP server runs with whatever access you grant it, and a compromised or malicious server can exfiltrate data or take unsanctioned actions. Treat every server as untrusted code: least-privilege scopes, human approval for irreversible actions, and no blanket production access. See the MCP security guide.