Cursor stop hook follow-up loop
When the agent loop ends, a stop hook can re-run checks and return followup_message to submit another turn automatically. loop_limit caps iterations (default 5).
proof: tests run countproof: logs blocks: agent stops before checks pass
By harness
| Harness | Status | How |
|---|---|---|
| Cursor | native | .cursor/hooks.json -> hooks.stop with a command. |
| Claude Code | adapted | Use a Stop hook with exit 2 instead. |
| Grok Build | adapted | Stop does not block in Grok Build; use PreToolUse on commit. |
| Codex | untested | Untested. |
| AGENTS.md generic | unsupported | Harness feature. |
By model family
| Claude | any Cursor model |
| GPT | any Cursor model |
| Grok | any Cursor model |
| Gemini | any Cursor model |
Sponsor slot: your verification tool here, one sponsor per week. Details
Proof of the week
One verification pattern every week, adapted per harness and model. No spam.