scar-code-review

Code review that learns from failures. Reflex arc blocks repeat mistakes without LLM calls. Combines systematic checklist review (security, performance, correctness, maintainability) with scar memory — when a review misses a bug, record a scar, and the reflex arc automatically flags similar patterns next time.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "scar-code-review" with this command: npx skills add aibenyclaude-coder/tetra-scar-code-review

scar-code-review

What this does

A code review system that learns from its own misses. Two layers work together:

  1. Checklist review — Regex/heuristic checks across 4 dimensions:

    • Security: SQL injection, hardcoded secrets, XSS, eval/exec
    • Performance: N+1 queries, missing pagination, unbounded SELECTs
    • Correctness: Unchecked nulls, off-by-one patterns, unhandled promises
    • Maintainability: Long functions, deep nesting, magic numbers
  2. Scar reflex arc — Pattern-matching against past review misses. When a review fails to catch a bug that later causes an incident, record a scar. Next time, the reflex fires before the LLM even looks at the diff.

No external dependencies. stdlib only. Python 3.9+.

Quick start

Review a file:

python3 scar_code_review.py review path/to/file.py

Check a diff against past scars:

python3 scar_code_review.py check-diff path/to/changes.diff

Record a missed review finding:

python3 scar_code_review.py record-miss \
  --what-missed "Missed SQL injection in user input handler" \
  --pattern "execute.*format.*user" \
  --severity critical

File format

JSONL, compatible with tetra-scar:

{"id":"rscar_1234","what_missed":"...","pattern":"...","severity":"critical","created_at":"..."}

Integration

from scar_code_review import review, reflex_check, record_miss, load_review_scars

# Review a file
findings = review("app/views.py")
for f in findings:
    print(f"{f['severity']} [{f['dimension']}] {f['message']} (line {f['line']})")

# Check diff against past scars
scars = load_review_scars()
blocks = reflex_check(diff_text, scars)
for b in blocks:
    print(f"BLOCKED: {b}")

# Record a miss after an incident
record_miss(
    what_missed="Missed unvalidated redirect",
    pattern="redirect.*request\\.GET",
    severity="high",
)

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

Agent Memory Loop

Lightweight self-improvement loop for AI agents. Capture errors, corrections, and discoveries in a fast one-line format, dedup them, queue recurring or criti...

Registry Source
3660Profile unavailable
Security

Cognitive Brain

Provides a cross-session AI memory and cognition system with four-layer memory, real-time sync, free thinking, intelligent prediction, and knowledge visualiz...

Registry SourceRecently Updated
8420Profile unavailable
Security

Engrm Sentinel

Use Engrm packs and Sentinel context to surface likely mistakes, risky patterns, and lessons that would have prevented them.

Registry Source
1450Profile unavailable
General

Super Brain

AI自我增强系统 - 让AI跨会话记住用户、持续进化。当需要长期记忆用户偏好、追踪对话历史、学习服务技巧、主动提供个性化服务时使用此技能。

Registry SourceRecently Updated
3390Profile unavailable