← Back to glossarytooling
MCP
Model Context Protocol is an open standard from Anthropic that defines how AI agents talk to tools, data sources, and APIs — like USB for AI.
Also known as: Model Context Protocol
In detail
Before MCP, every platform invented its own tool-calling interfaces. MCP standardizes it: a server exposes tools (functions), resources (data), and prompts; a client (your agent — any platform) consumes them.
Benefit for you: skills we build to MCP work not just in Agent Hub, but also in Cursor, Claude Desktop, Continue, or any other MCP-capable client. No vendor lock-in.
Example
We build an MCP server for your CRM. The agent in Agent Hub uses it, but your developers can also wire it into Claude Desktop directly to query CRM data on demand.
Related terms
- StreamingStreaming means the language model's response is delivered token by token as soon as it's generated — the user sees the beginning while the model is still writing.
- AI agentAn AI agent is a program built on a language model that completes tasks on its own: it understands a request, plans steps, calls tools, and responds with a result instead of just text.
- Context windowThe context window is the maximum amount of text (measured in tokens) a language model can process at once — typically 128k to 1M tokens with current models.