Agent complete (stop)
Hook that fires when the agent finishes a turn — implement quality gates, force continuation, or trigger follow-up workflows
Observation Hooks Hooks & Lifecycle
Supported
Partial
No
Unknown
2.1.72 - 2.1.143
3.82.0 - 3.83.0
0.114.0 - 0.128.0
2026.03 - 2026.05
2.5 - 3.2
1.13.3 - 2.1.29
2.1.10 - 2.1.72
3.82.0
0.88.0 - 0.113.0
2025.10 - 2026.03
2.4 - 2.5
1.13.3
1.0.38 - 2.1.10
3.82.0
0.2.0 - 0.88.0
2025.06 - 2025.10
1.0 - 2.0
1.2.1 - 1.13.3
0.2.54 - 1.0.0
3.36.0 - 3.82.0
—
2025.06
0.45 - 1.0
1.2.1
0.2.31 - 0.2.54
1.5.0 - 3.33.0
—
—
0.45
1.0.2 - 1.2.1
Supported by 4 of 6 agents.
HoverTap any cell for version details.
Agent event mapping
Sub-capabilities
- Force the agent to keep working by returning a continuation signal
- Implement quality checks before accepting the agent's work
- Claude Code: StopFailure fires on API errors
- Cursor: followup_message for auto-continue with configurable loop_limit
- Cline: TaskComplete on success, TaskCancel when user stops a running task