AI agent
An LLM-powered system that takes actions — calling tools, reading data, completing multi-step tasks — rather than only generating text.
An AI agent is a model wrapped in a loop that can do things: it decides which tool to call, calls it, reads the result, and keeps going until a task is done. The difference from a plain chatbot is action — an agent can search the web, run code, query a database, or hit an API, often via the Model Context Protocol.
Agents change how tools get discovered. When the agent, not the human, picks which service to use, the question becomes: is your tool reachable and legible to it? That's why agent-readiness and MCP support increasingly decide whether a product shows up in real workflows.