An agent that decides its own next step needs a limit that is not its own judgement. There are four, and a run stops on the first one it reaches.
The four ceilings
Steps
How many actions a run may take. The bluntest limit and usually the most useful.
Spend
Measured from the usage log rather than estimated, so it reflects what was actually billed.
Tokens
Total tokens across the run.
Deadline
Wall-clock. A run that is still going when it passes stops there.
Zero means no limit
A ceiling of zero means no limit of that kind. For an agent that decides its own next step, that is rarely what anyone wants — set at least a step count and a spend limit before running anything unattended.
Frozen when the run starts
The ceilings are written onto the run at the moment it begins. Editing a playbook or a global setting between clicking and the queue picking the job up cannot change a run already authorised — what you approved is what runs.
Reading what stopped a run
Agent Runs shows the reason: the step count, spend, tokens or the deadline. A run that stopped on a ceiling is not a failure — it did what it was allowed to do and then stopped, which is the point.
Did this answer it? Tell us if not and we will fix the article.