// Post-implementation review/fix loop. Use host-appropriate review coverage on your implementation, fix issues, iterate until highly confident, then commit.
Use for complex or high-risk implementation work that needs plan-first execution, explicit verification design, pre-code review, and autonomous post-implementation hardening through self-test, simplify, and review-loop.
Prove the requested goal works on the highest-signal affected surface before closing. Default closing behavior for every implementation task — not just when explicitly asked. Also triggers on: 'make sure it works', 'test it yourself', 'don't just tell me it's done', 'I don't want to test this'.
Lightweight read-only audit. Use host-appropriate review coverage for evidence and findings. Iterate on disagreement or low confidence. No code changes.
Iteratively clarify requirements for a new task until reaching 90%+ confidence, then auto-invoke plan-loop.
Orchestrate complex tasks with agent teams. Clarify requirements, plan with dual-model review, implement with TeamCreate, simplify, review, and ship.
Audit agent skills in a repo against best practices — checks .claude/skills/ and .agents/skills/ for quality, structure, and coverage gaps. Use when the user wants to review, improve, or evaluate their skills, or when they mention skill quality, skill audit, or skill best practices.
| name | review-loop |
| description | Post-implementation review/fix loop. Use host-appropriate review coverage on your implementation, fix issues, iterate until highly confident, then commit. |
After implementation, harden it with review coverage appropriate to the current host. Fix, iterate, commit.
Read-only analysis → use audit-loop instead.
Planning before implementation → use plan-loop instead.
Read ~/dev/agent-guards/AGENTS.md before applying fixes — the safety, workflow, and code conventions that govern all agent work.
references/review-protocol.md. Spin up all review agents in parallel using the host-specific coverage rules. Pass relevant context (what you built, why, what changed, how you verified). Pre-warm any deferred tools only when the selected host path actually needs them.self-test to re-verify on the real surface — don't trust code review alone.self-test to re-verify — don't trust the code review alone to confirm the fix works on the real surface. Skipping this is how "reviewed and fixed" changes still ship broken.After each round, assess confidence 0-100 with top unknowns and what must be true to close:
<70: another round, mandatory.70-84: one more focused round on residual risks. If still 70-84 after that, escalate to user with findings.>=85: eligible to close. Auto commit, push, and create PR.