iterate-review — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited iterate-review (Agent Skill) and scored it 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You are in AUTONOMOUS MODE. Do NOT ask questions. Analyze and fix.
TARGET: $ARGUMENTS
============================================================ PRE-REVIEW: STATIC VALIDATION ============================================================
Before reviewing code, run automated checks to clear the noise:
flutter analyze. Fix all errors and warnings.dart fix --apply.tsc --noEmit. Fix type errors.dart:io imports in web-reachable code.sed portability, path breakage after reorg):
bash -n on all .sh files to catch syntax errors (stray fi, unmatched brackets).sed -i usage — needs '' arg on macOS, not on Linux. Flag unportable usage.$SCRIPT_DIR, ../, ./ path referencesstill resolve correctly.
$CALLER_ID vs $AGENT_ID).Fix all issues. Commit: "fix: pre-review static validation" If clean, skip commit.
============================================================ PROCESS (max 5 iterations) ============================================================
=== EACH ITERATION ===
but are never called from the client? (e.g., validatePayment exists but client does client-only checks.) This is a CRITICAL issue.
are missing from the client model's fromMap/toMap? (e.g., Cloud Function writes warningCount but model doesn't deserialize it.)
no matching rule in firestore.rules? Or rules that are too permissive?
pending period) actually passed through to the functions that use them, or are hardcoded defaults used instead?
or exceeding 600 lines. Recommend splitting into domain-specific modules.
should be wrapped in try/catch with user-visible error feedback.
Run the /analyze skill scoped to the code under review. Include all analysis phases: consistency audit, server-side validation wiring, model-to-Cloud-Function field completeness, Firebase rules, and platform compatibility. If analysis finds Critical or Warning issues, add them to the fix list for the next iteration. Critical issues take priority over code review issues.
analysis shows zero Critical/Warning issues.
=== ITERATION FOCUS ===
=== DO NOT ===
=== COMMIT DISCIPLINE (learned from recall analysis) ===
as the related code fix.
=== OUTPUT ===
Brief summary after each iteration:
Final summary:
NEXT STEPS:
/qa to verify everything works end-to-end."/arch-review for architect-level structural review."/readme to update project documentation with the current state."/ux to audit accessibility and design standards."============================================================ SELF-HEALING VALIDATION (max 3 iterations) ============================================================
After completing fixes, re-validate your work:
STOP when:
IF STILL FAILING after 3 iterations:
============================================================ SELF-EVOLUTION TELEMETRY ============================================================
After producing output, record execution metadata for the /evolve pipeline.
Check if a project memory directory exists:
~/.claude/projects/skill-telemetry.md in that memory directoryEntry format:
### /iterate-review — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.