How-to guide
What is an AI agent?
An agent is an AI that works toward a goal by planning, using tools and checking its own results. Learn what that means and what agents are not.
- Time 8 min
- Level Beginner
- Steps 5
- Updated
A chatbot answers a question. An AI agent is given a goal, then decides what to do next, uses tools, looks at the results and keeps going until the job is done or it needs help.
Step by step
0/5 done-
1 The four parts of an agent
- A model that reasons and decides.
- Instructions: its role, rules and style (the system prompt).
- Tools: ways to act, such as reading files, running commands, searching the web or calling an app.
- A loop: act, observe, decide again.
-
2 What an agent is not
It is not magic and not always the right choice. A single, well-written prompt is cheaper and more predictable for simple tasks. Reach for an agent when the path to the answer is not known in advance.
-
3 Examples
A code reviewer that reads a diff and the files around it. A research assistant that searches, reads and cites. A support triager that classifies messages and drafts replies.
-
4 Where skills fit
An agent can load skills for specific jobs. The agent is the worker; skills are the how-to manuals it can open.
-
5 Where to go next
Read How to create a subagent in Claude Code and Write a good agent system prompt, then browse the agents library.
Was this helpful?
Thanks — that helps us improve this page.