Glossary

Every concept in this sheet, plus the smaller terms it uses. Select a concept to open its card.

Capabilities: Declarations of optional features each side supports, such as tools with listChanged on a server or elicitation on a client.
Client: The connector inside the host that talks to exactly one server.

Host: The AI application the user interacts with, such as a desktop chat app, an IDE or an agent.

Model Context Protocol: An open protocol that defines how AI applications discover and use external tools, data and prompt templates, using JSON-RPC 2.0 messages over a transport.
Multi round-trip requests: When a server needs more from the client mid-request (user input, for example), it returns a result with resultType: "input_required" and the questions.

Notifications and subscriptions: One-way messages such as progress updates or notifications/tools/list_changed.

Protocol errors and tool errors: Protocol problems (unknown tool, malformed params) are JSON-RPC errors.
Protocol versions and extensions: The specification is versioned by date, for example 2025-11-25 and 2026-07-28.

Server: A program that exposes tools, resources and prompts over MCP.
server/discover and version negotiation: Servers implement server/discover to advertise their supported versions, capabilities and identity.
Stateless requests and _meta: Since 2026-07-28, each request carries its protocol version and client capabilities (and usually client info) in _meta.

The initialize handshake (older revisions): Up to 2025-11-25, a connection began with an initialize request exchanging versions and capabilities, then an initialized notification.
The M times N problem: Without a standard, M AI applications and N services need M times N custom integrations.
Tools, resources and prompts: Tools are functions the model can call; resources are data the application can read into context; prompts are templates the user picks.