Glossary
Agent loop
What is an agent loop?
The agent loop is the cycle an AI agent runs to do a task: the harness gives the model the context, the model chooses the next step, the harness carries it out and feeds back the result, and the cycle repeats until the task is done or a stop rule applies.
Also called: Agentic loop
Why it matters
Most of what makes an agent reliable happens around this loop: what the model sees at each turn, which actions it may take, how each result is checked, and when it must stop or ask a person.
Example
To handle a refund request, an agent loops through reading the ticket, looking up the order and checking the refund policy, then drafts a reply and stops so a person can approve it.