Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyzes code diffs and files to identify bugs, security vulnerabilities, code smells, and architectural concerns, then produces a structured review report with prioritized, actionable feedback.
.claude/skills/paperclipai-code-reviewer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✓→✗ | ▼ Worse | -48% | 0% |
| case-10 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | -2% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 19% | 0% |
| case-08 | ✓→✓ | ＝ Same ✓ | -2% | 0% |
name: code-reviewer description: > Analyzes code diffs and files to identify bugs, security vulnerabilities, code smells, and architectural concerns, then produces a structured review report with prioritized, actionable feedback. metadata: sources:
repo: jeffallan/claude-skills path: skills/code-reviewer/SKILL.md commit: 3bf9a24b76a7c122f1fc05e83929fbc84e1c207a attribution: Jeffallan license: MIT usage: referenced
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 15,498 | 12,155 | -22% | 1 | 1 | 0% | 1,516 | 1,427 | -6% | 0 | 0 | — |
case-02 | fail→fail | 15,148 | 16,858 | +11% | 1 | 1 | 0% | 1,807 | 2,031 | +12% | 0 | 0 | — |
case-03 | fail→fail | 6,311 | 12,666 | +101% | 1 | 1 | 0% | 898 | 1,951 | +117% | 0 | 0 | — |
case-04 | fail→fail | 25,335 | 24,466 | -3% | 1 | 1 | 0% | 6,177 | 6,326 | +2% | 0 | 0 | — |
case-10 | pass→pass | 9,827 | 8,919 | -9% | 1 | 1 | 0% | 1,817 | 1,901 | +5% | 0 | 0 | — |
case-05 | fail→fail | 15,223 | 15,335 | +1% | 1 | 1 | 0% | 3,314 | 3,516 | +6% | 0 | 0 | — |
case-06 | pass→pass | 7,270 | 7,247 | -0% | 1 | 1 | 0% | 1,357 | 1,326 | -2% | 0 | 0 | — |
case-07 | pass→pass | 12,875 | 11,496 | -11% | 1 | 1 | 0% | 2,022 | 2,405 | +19% | 0 | 0 | — |
case-08 | pass→pass | 8,187 | 7,907 | -3% | 1 | 1 | 0% | 1,526 | 1,488 | -2% | 0 | 0 | — |
case-09 | pass→pass | 11,602 | 9,155 | -21% | 1 | 1 | 0% | 1,866 | 1,856 | -1% | 0 | 0 | — |
case-11 | pass→pass | 9,909 | 10,275 | +4% | 1 | 1 | 0% | 1,935 | 1,749 | -10% | 0 | 0 | — |
case-12 | pass→pass | 8,619 | 11,320 | +31% | 1 | 1 | 0% | 1,537 | 1,872 | +22% | 0 | 0 | — |
case-13 | pass→pass | 8,513 | 6,640 | -22% | 1 | 1 | 0% | 1,536 | 1,441 | -6% | 0 | 0 | — |
case-14 | pass→pass | 7,385 | 5,004 | -32% | 1 | 1 | 0% | 1,162 | 1,166 | +0% | 0 | 0 | — |
case-15 | pass→pass | 10,505 | 8,631 | -18% | 1 | 1 | 0% | 1,865 | 1,808 | -3% | 0 | 0 | — |
case-16 | pass→pass | 9,100 | 9,700 | +7% | 1 | 1 | 0% | 1,746 | 1,718 | -2% | 0 | 0 | — |
case-17 | pass→pass | 8,350 | 7,316 | -12% | 1 | 1 | 0% | 1,556 | 1,552 | -0% | 0 | 0 | — |
case-18 | pass→fail | 16,899 | 8,861 | -48% | 1 | 1 | 0% | 2,482 | 1,289 | -48% | 0 | 0 | — |
case-19 | pass→pass | 9,135 | 7,528 | -18% | 1 | 1 | 0% | 1,748 | 1,296 | -26% | 0 | 0 | — |
case-20 | pass→pass | 11,243 | 8,796 | -22% | 1 | 1 | 0% | 2,029 | 1,412 | -30% | 0 | 0 | — |
case-21 | pass→pass | 9,033 | 8,039 | -11% | 1 | 1 | 0% | 1,628 | 1,621 | -0% | 0 | 0 | — |
case-22 | pass→pass | 11,163 | 13,434 | +20% | 1 | 1 | 0% | 2,091 | 2,172 | +4% | 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 -100 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.