Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when asked to review MERN stack code - comprehensive code reviewer that checks project health, security, maintainability, performance, testing, and architecture. Combines general code quality analysis with MERN-specific expertise.
.claude/skills/aiskillstore-code-reviewer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 50% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 95% | 0% |
| case-01 | ✓→✗ | ▼ Worse | -34% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 99% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 99% | 0% |
resources/implementation-playbook.md.You are an elite code review expert specializing in modern code analysis techniques, AI-powered review tools, and production-grade quality assurance.
Master code reviewer focused on ensuring code quality, security, performance, and maintainability using cutting-edge analysis tools and techniques. Combines deep technical expertise with modern AI-assisted review processes, static analysis tools, and production reliability practices to deliver comprehensive code assessments that prevent bugs, security vulnerabilities, and production incidents.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→fail | 22,629 | 14,376 | -36% | 1 | 1 | 0% | 4,356 | 2,870 | -34% | 0 | 0 | — |
case-02 | fail→fail | 28,684 | 35,166 | +23% | 1 | 1 | 0% | 6,374 | 8,707 | +37% | 0 | 0 | — |
case-03 | fail→pass | 9,684 | 5,754 | -41% | 1 | 1 | 0% | 1,523 | 2,278 | +50% | 0 | 0 | — |
case-04 | fail→fail | 18,023 | 22,814 | +27% | 1 | 1 | 0% | 3,161 | 5,394 | +71% | 0 | 0 | — |
case-05 | pass→pass | 10,104 | 11,325 | +12% | 1 | 1 | 0% | 1,807 | 3,596 | +99% | 0 | 0 | — |
case-06 | pass→pass | 11,084 | 12,042 | +9% | 1 | 1 | 0% | 1,849 | 3,686 | +99% | 0 | 0 | — |
case-19 | pass→pass | 73,523 | 15,699 | -79% | 1 | 1 | 0% | 1,945 | 4,225 | +117% | 0 | 0 | — |
case-07 | pass→pass | 12,315 | 13,004 | +6% | 1 | 1 | 0% | 2,099 | 3,794 | +81% | 0 | 0 | — |
case-08 | pass→pass | 13,987 | 25,084 | +79% | 1 | 1 | 0% | 2,501 | 4,349 | +74% | 0 | 0 | — |
case-09 | pass→pass | 16,009 | 11,235 | -30% | 1 | 1 | 0% | 1,364 | 3,479 | +155% | 0 | 0 | — |
case-10 | pass→pass | 13,013 | 16,143 | +24% | 1 | 1 | 0% | 2,278 | 4,382 | +92% | 0 | 0 | — |
case-11 | pass→pass | 10,858 | 11,888 | +9% | 1 | 1 | 0% | 1,871 | 3,816 | +104% | 0 | 0 | — |
case-12 | pass→pass | 10,607 | 17,659 | +66% | 1 | 1 | 0% | 1,920 | 4,383 | +128% | 0 | 0 | — |
case-13 | pass→pass | 16,610 | 23,205 | +40% | 1 | 1 | 0% | 1,214 | 3,575 | +194% | 0 | 0 | — |
case-14 | pass→pass | 11,240 | 17,763 | +58% | 1 | 1 | 0% | 2,083 | 4,180 | +101% | 0 | 0 | — |
case-15 | pass→pass | 10,709 | 9,411 | -12% | 1 | 1 | 0% | 1,562 | 3,155 | +102% | 0 | 0 | — |
case-16 | fail→pass | 15,855 | 13,370 | -16% | 1 | 1 | 0% | 2,098 | 4,092 | +95% | 0 | 0 | — |
case-17 | pass→pass | 14,097 | 15,178 | +8% | 1 | 1 | 0% | 1,877 | 3,830 | +104% | 0 | 0 | — |
case-18 | pass→pass | 10,385 | 25,339 | +144% | 1 | 1 | 0% | 1,836 | 3,755 | +105% | 0 | 0 | — |
case-20 | pass→pass | 12,105 | 14,055 | +16% | 1 | 1 | 0% | 2,073 | 3,537 | +71% | 0 | 0 | — |
case-21 | pass→pass | 15,954 | 11,324 | -29% | 1 | 1 | 0% | 2,254 | 3,412 | +51% | 0 | 0 | — |
case-22 | fail→fail | 11,435 | 11,230 | -2% | 1 | 1 | 0% | 1,807 | 3,555 | +97% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/21/2026 | -60% |
Other measured skills in the registry, with their headline benchmark lift.