Model Context Protocol (MCP)

What it is

The MCP project describes it as a USB-C port for AI applications: one standard plug between an assistant and the systems it needs. A business app offers an MCP server; an assistant or agent acts as the MCP client.

How it works

The server describes the tools and data it offers (for example “search contacts” or “create invoice”). The client shows those descriptions to the model, which decides when to call them. The client then runs the call and passes the result back. Our Agent-ready page lists which tools we have confirmed offer an official MCP server, with the source for each.

Benefits

• Build once, connect everywhere. One server works with every compatible client.
• Live data, real actions. The assistant reads current records and can act, not just draft text.
• Less glue. Fewer hand-built connectors and automations between your AI tool and your stack.

Limits and risks

• Context cost. Every connected server’s tool descriptions take up room in the prompt. Simon Willison suggests command-line tools as a leaner option when the agent has a shell (see MCP or CLI).
• Permissions. A connected agent can do whatever the connection allows. Scope access before you connect (see letting an agent act safely).
• Unofficial servers. Community-built servers vary in quality and security. We only mark MCP as “yes” with an official source.