Review Fix Loop
Testing & Qualityreview-fix-loopReview Fix Loop is an Agent Skill for testing & quality published in the spacedriveapp/spacebot repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
This skill should be used when the user asks to "address review feedback", "fix PR comments", "close findings", "respond to reviewer notes", or "reduce review churn". Enforces a finding-to-evidence closure loop for every review round.
QUICK START
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/spacedriveapp/spacebot/blob/HEAD/.agents/skills/review-fix-loop/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/review-fix-loop, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
Review Fix Loop
Goal
Close review rounds with minimal back-and-forth by mapping every finding to a code change and proof.
Execution Loop
- Parse findings into
P1,P2,P3. - Build a closure table before editing.
- Implement the smallest coherent fix batch.
- Run targeted verification.
- Update closure table with outcomes.
- Run broader gate checks.
Closure Table
Use one row per finding:
| Finding | Severity | Owned files | Planned change | Verification command | Result |
|---|
Re-Run Control
- If the same command fails twice, stop rerunning.
- Isolate a smaller reproduction command.
- Patch the smallest likely cause.
- Re-run narrow verification before broad checks.
Verification Ladder
- Start narrow: unit/module behavior checks for touched paths.
- Continue medium: compile/lint/type checks for touched surfaces.
- End broad: repository gate commands.
Handoff Requirements
- Findings closed
- Commands executed
- Pass/fail evidence per finding
- Remaining open findings with rationale
- Residual risk