← ClaudeAtlas

code-review-guidelisted

Review code for quality, security, performance, and architecture compliance with a 100-point scoring framework and verdicts (APPROVED/APPROVED_WITH_NOTES/NEEDS_WORK). Trigger on "review this code/my changes", "check for issues", "score the code", "is this ready to ship", "PR review", "pre-merge review", "second pair of eyes", "honest score and verdict", "before I open the PR", "check for N+1 queries", "missing error handling", "input validation", "code smells", "review every file they touched". EVALUATES code quality with structured findings + scoring — not for verifying stubs/wiring (use verification-patterns), not for searching existing code (use codebase-scan), not for assessing task complexity (use complexity-assessment).
emrecdr/devt · ★ 0 · Code & Development · score 75
Install: claude install-skill emrecdr/devt
# Code Review Guide ## Overview A code review is a structured inspection that produces a scored assessment with specific, actionable findings. Every finding must reference a concrete location, describe the problem, explain the impact, and suggest a fix. Reviews are objective. They evaluate code against documented project standards, not personal preferences. If a standard does not exist for something, it is not a finding. **Precision mandate**: Every finding must name a specific file, line number, and exact violation. If a finding could apply to any codebase, it is too vague to report. ## When NOT to Use Skip for self-review during implementation — this is for the formal review phase. ## Time Budget - **Quick review** (1-3 files): 2-3 minutes - **Full review** (5+ files): 5-10 minutes ## The Iron Law ``` EVERY VALID FINDING MUST BE REPORTED — NO FILTERING, NO MERCY ``` An unreported finding is a silently disabled quality gate. If the code violates a documented standard, report it — regardless of origin, perceived severity, or whether the developer "probably knows." The tendency to filter findings by perceived importance, origin ("pre-existing"), or social comfort ("it's minor") directly undermines the review's value. Users rely on the review to catch what they missed. A finding omitted because it seemed small is a finding the developer never gets to evaluate. The reviewer's job is detection, not triage — the developer decides what to fix. ## The Process ### Step