A model answers. An agent acts.
The difference between an AI model, which gives you text, and an agent, which takes steps to reach a goal.
What AI agents are, how they work and how to use them well. No code.
For anyone who wants to understand AI agents, with no technical background needed. Each chapter explains one idea in plain words and lets you try it in a simulation. At the end you can tell what an agent can and cannot do, spot the main risks and decide what to hand to one.
The difference between an AI model, which gives you text, and an agent, which takes steps to reach a goal.
Every agent runs the same loop: it looks at what it knows, decides the next step, acts, and looks again.
Tools are how an agent looks things up and gets things done, and the model knows each tool only by its label.
What an agent can see at each step is limited, so what it keeps, drops and writes down decides what it knows.
Big jobs are split into steps, and sometimes shared among several agents led by one.
Standards are agreed message formats that let agents, tools, apps and shops built by different people work together.
Agents read things they did not write and take actions with effects, so they need limits and a person's approval at the right moments.
Choosing what to hand to an agent, briefing it clearly, and checking what it did.