AI Glossary · Last reviewed August 2026

AI Agents

Hand-written by a real person. Reviewed against current practice in August 2026.

Definition

A program that takes goals and figures out the steps to reach them, calling external tools to act in the world.

Why it matters

AI agents go beyond simple chatbots - they can plan multi-step tasks, use external tools, and act on your behalf. Understanding agents helps you decide whether a tool just answers questions or can actually get work done for you.

As agents become more capable, knowing their limitations - like hallucination risks and permission boundaries - helps you use them safely and effectively.

How it works

4 steps

STEP 01

Receive a goal

The user gives the agent a high-level objective like book a flight or debug this code.

STEP 02

Plan the steps

The agent breaks the goal into smaller sub-tasks and decides which tools or APIs it needs to call.

STEP 03

Execute and observe

It runs each step, reads the result, and adjusts its plan if something unexpected happens.

STEP 04

Return the result

Once the goal is met, the agent delivers the final output back to the user.

Related terms

From the glossary

Frequently asked questions

Are AI agents reliable enough to use in production?+

It depends on the task and the safety guardrails in place. Narrow, well-defined tasks with human checkpoints work well today. Open-ended agents with broad permissions are still maturing.

New to AI Agents?

See the tools that use it.

The fastest way to understand AI Agents is to see it inside real products. Browse hand-reviewed tools that put it to work, each one checked by a person before it was listed.

Browse hand-reviewed AI tools