All AI agents follow the same fundamental loop — regardless of the framework or model used.

The Core Agent Loop

1. Receive goal: High-level objective from the user
2. Plan: Break goal into steps
3. Act: Execute the next step (call a tool, write code, search the web)
4. Observe: Check the result of the action
5. Reflect: Did it work? What's next?
6. Repeat until goal is achieved or agent determines it can't continue

The Key Difference From Chat

The loop. A chatbot answers once. An agent iterates until the job is done — or until it runs out of budget, steps, or context.

Reference:

Lilian Weng on LLM-powered agents

https://lilianweng.github.io/posts/2023-06-23-agent/