Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →You are a code refactoring expert specializing in clean code principles, SOLID design patterns, and modern software engineering best practices. Analyze and refactor the provided code to improve its quality, maintainability, and performance.
.claude/skills/dokhacgiakhoa-code-refactoring-refactor-clean/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 277% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 6% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 22% | 0% |
You are a code refactoring expert specializing in clean code principles, SOLID design patterns, and modern software engineering best practices. Analyze and refactor the provided code to improve its quality, maintainability, and performance.
The user needs help refactoring code to make it cleaner, more maintainable, and aligned with best practices. Focus on practical improvements that enhance code quality without over-engineering.
$ARGUMENTS
resources/implementation-playbook.md.resources/implementation-playbook.md for detailed patterns and examples.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | pass→pass | 13,422 | 11,315 | -16% | 1 | 1 | 0% | 2,087 | 2,158 | +3% | 0 | 0 | — |
case-01 | pass→pass | 2,374 | 8,867 | +274% | 1 | 1 | 0% | 368 | 747 | +103% | 0 | 0 | — |
case-02 | pass→pass | 11,259 | 3,645 | -68% | 1 | 1 | 0% | 1,911 | 921 | -52% | 0 | 0 | — |
case-03 | pass→fail | 13,813 | 13,407 | -3% | 1 | 1 | 0% | 3,027 | 3,206 | +6% | 0 | 0 | — |
case-04 | pass→pass | 12,674 | 15,791 | +25% | 1 | 1 | 0% | 2,277 | 3,649 | +60% | 0 | 0 | — |
case-05 | pass→pass | 14,431 | 17,269 | +20% | 1 | 1 | 0% | 3,059 | 4,060 | +33% | 0 | 0 | — |
case-06 | fail→pass | 5,158 | 15,135 | +193% | 1 | 1 | 0% | 918 | 3,458 | +277% | 0 | 0 | — |
case-07 | fail→pass | 14,404 | 22,946 | +59% | 1 | 1 | 0% | 3,160 | 4,064 | +29% | 0 | 0 | — |
case-08 | pass→pass | 10,733 | 9,228 | -14% | 1 | 1 | 0% | 1,704 | 1,932 | +13% | 0 | 0 | — |
case-09 | pass→pass | 14,908 | 15,718 | +5% | 1 | 1 | 0% | 2,424 | 2,854 | +18% | 0 | 0 | — |
case-10 | pass→fail | 8,133 | 7,449 | -8% | 1 | 1 | 0% | 1,202 | 1,462 | +22% | 0 | 0 | — |
case-11 | pass→pass | 14,166 | 15,503 | +9% | 1 | 1 | 0% | 2,766 | 3,259 | +18% | 0 | 0 | — |
case-12 | pass→pass | 14,268 | 14,663 | +3% | 1 | 1 | 0% | 2,521 | 3,002 | +19% | 0 | 0 | — |
case-13 | pass→pass | 9,643 | 9,426 | -2% | 1 | 1 | 0% | 1,836 | 2,085 | +14% | 0 | 0 | — |
case-14 | fail→pass | 4,840 | 3,405 | -30% | 1 | 1 | 0% | 790 | 799 | +1% | 0 | 0 | — |
case-15 | pass→pass | 13,029 | 13,730 | +5% | 1 | 1 | 0% | 2,152 | 3,014 | +40% | 0 | 0 | — |
case-16 | pass→pass | 9,970 | 10,482 | +5% | 1 | 1 | 0% | 1,767 | 2,321 | +31% | 0 | 0 | — |
case-17 | pass→pass | 11,065 | 9,718 | -12% | 1 | 1 | 0% | 2,168 | 2,129 | -2% | 0 | 0 | — |
case-19 | pass→pass | 16,096 | 10,615 | -34% | 1 | 1 | 0% | 2,206 | 2,482 | +13% | 0 | 0 | — |
case-20 | pass→pass | 17,127 | 10,469 | -39% | 1 | 1 | 0% | 2,546 | 2,378 | -7% | 0 | 0 | — |
case-21 | pass→pass | 8,703 | 10,092 | +16% | 1 | 1 | 0% | 1,433 | 1,890 | +32% | 0 | 0 | — |
case-22 | pass→pass | 12,958 | 12,946 | -0% | 1 | 1 | 0% | 2,253 | 2,168 | -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 +5 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.
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.
Other measured skills in the registry, with their headline benchmark lift.