Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Multi-language code coverage analysis, reporting, and quality gate enforcement
.claude/skills/a5c-ai-code-coverage-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -63% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -59% | 0% |
This skill provides expert-level capabilities for code coverage analysis across multiple languages, enabling coverage collection, report generation, and quality gate enforcement.
automation-framework.js - Framework coverage setupmutation-testing.js - Coverage for mutation testingquality-gates.js - Coverage-based gatescontinuous-testing.js - CI/CD coveragenyc / c8 - JavaScript coveragecoverage.py - Python coverageJaCoCo - Java coveragejavascript{ kind: 'skill', skill: { name: 'code-coverage', context: { action: 'analyze', language: 'javascript', reportFormats: ['html', 'lcov', 'json'], thresholds: { lines: 80, branches: 75, functions: 80 } } } }
The skill auto-detects project language and configures appropriate coverage tools.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 29,636 | 29,168 | -2% | 1 | 1 | 0% | 4,927 | 5,766 | +17% | 0 | 0 | — |
case-02 | fail→fail | 12,925 | 12,426 | -4% | 1 | 1 | 0% | 2,280 | 2,688 | +18% | 0 | 0 | — |
case-03 | pass→pass | 6,247 | 7,403 | +19% | 1 | 1 | 0% | 1,132 | 1,696 | +50% | 0 | 0 | — |
case-04 | pass→pass | 14,353 | 14,112 | -2% | 1 | 1 | 0% | 2,724 | 3,297 | +21% | 0 | 0 | — |
case-05 | fail→pass | 18,071 | 19,602 | +8% | 1 | 1 | 0% | 2,629 | 3,289 | +25% | 0 | 0 | — |
case-06 | pass→pass | 15,455 | 14,800 | -4% | 1 | 1 | 0% | 2,595 | 3,083 | +19% | 0 | 0 | — |
case-07 | pass→pass | 17,299 | 14,045 | -19% | 1 | 1 | 0% | 2,740 | 2,887 | +5% | 0 | 0 | — |
case-08 | pass→pass | 16,985 | 22,486 | +32% | 1 | 1 | 0% | 2,680 | 3,898 | +45% | 0 | 0 | — |
case-09 | pass→pass | 35,720 | 14,745 | -59% | 1 | 1 | 0% | 2,498 | 2,883 | +15% | 0 | 0 | — |
case-10 | fail→pass | 13,632 | 4,526 | -67% | 1 | 1 | 0% | 2,295 | 1,189 | -48% | 0 | 0 | — |
case-11 | fail→pass | 11,948 | 2,450 | -79% | 1 | 1 | 0% | 2,041 | 756 | -63% | 0 | 0 | — |
case-12 | fail→pass | 18,727 | 6,033 | -68% | 1 | 1 | 0% | 2,795 | 1,301 | -53% | 0 | 0 | — |
case-13 | fail→pass | 10,273 | 1,880 | -82% | 1 | 1 | 0% | 1,784 | 739 | -59% | 0 | 0 | — |
case-14 | fail→pass | 3,765 | 2,684 | -29% | 1 | 1 | 0% | 836 | 977 | +17% | 0 | 0 | — |
case-15 | pass→pass | 15,321 | 11,406 | -26% | 1 | 1 | 0% | 2,259 | 2,535 | +12% | 0 | 0 | — |
case-16 | pass→pass | 15,683 | 13,923 | -11% | 1 | 1 | 0% | 2,923 | 3,160 | +8% | 0 | 0 | — |
case-17 | pass→pass | 11,913 | 12,320 | +3% | 1 | 1 | 0% | 1,932 | 2,440 | +26% | 0 | 0 | — |
case-18 | pass→pass | 10,625 | 10,441 | -2% | 1 | 1 | 0% | 1,747 | 2,133 | +22% | 0 | 0 | — |
case-19 | pass→pass | 12,626 | 13,824 | +9% | 1 | 1 | 0% | 2,278 | 2,890 | +27% | 0 | 0 | — |
case-20 | pass→pass | 13,741 | 13,978 | +2% | 1 | 1 | 0% | 2,459 | 3,309 | +35% | 0 | 0 | — |
case-21 | pass→pass | 9,691 | 7,498 | -23% | 1 | 1 | 0% | 1,873 | 2,002 | +7% | 0 | 0 | — |
case-22 | pass→pass | 17,549 | 17,761 | +1% | 1 | 1 | 0% | 3,102 | 3,705 | +19% | 0 | 0 | — |
case-23 | pass→pass | 14,514 | 13,253 | -9% | 1 | 1 | 0% | 2,693 | 2,996 | +11% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 comparable cases.
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.