Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Detects the test framework, runs the suite, collects coverage data, identifies failures with root cause analysis, and reports results in a structured format. Diagnoses each failure by reading both the test and the code under test.
.claude/skills/miosa-osa-test/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 660% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 403% | 0% |
/test scope] --type unit|integration|e2e|all] --coverage]
Run test suite with optional scope filtering and coverage reporting.
| Arg | Type | Required | Description | |-----|------|----------|-------------| | scope | string | No | File, directory, or module to test. Default: all | | --type | string | No | Test type filter. Default: all | | --coverage | flag | No | Generate coverage report |
Genre: report Format: Markdown test results + coverage summary
Produces:
1. Run specified test suite
2. Collect results and coverage data
3. QA engineer evaluates against quality thresholds
4. Generate structured report
5. Flag any coverage decreases from baseline/test
/test src/services/ --type unit --coverage
/test --type integration
/test src/api/users.test.ts| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 9,132 | 7,464 | -18% | 1 | 1 | 0% | 1,708 | 1,729 | +1% | 0 | 0 | — |
case-01 | fail→pass | 2,759 | 12,682 | +360% | 1 | 1 | 0% | 302 | 2,295 | +660% | 0 | 0 | — |
case-02 | fail→pass | 23,854 | 9,349 | -61% | 1 | 1 | 0% | 3,664 | 2,055 | -44% | 0 | 0 | — |
case-03 | pass→pass | 11,363 | 8,658 | -24% | 1 | 1 | 0% | 1,590 | 1,946 | +22% | 0 | 0 | — |
case-04 | pass→pass | 12,480 | 12,594 | +1% | 1 | 1 | 0% | 2,495 | 2,784 | +12% | 0 | 0 | — |
case-06 | pass→pass | 10,421 | 8,923 | -14% | 1 | 1 | 0% | 1,622 | 1,739 | +7% | 0 | 0 | — |
case-07 | pass→pass | 15,100 | 8,641 | -43% | 1 | 1 | 0% | 2,872 | 1,875 | -35% | 0 | 0 | — |
case-08 | fail→pass | 9,105 | 8,287 | -9% | 1 | 1 | 0% | 1,567 | 1,814 | +16% | 0 | 0 | — |
case-09 | fail→pass | 10,947 | 7,478 | -32% | 1 | 1 | 0% | 1,915 | 1,607 | -16% | 0 | 0 | — |
case-10 | fail→pass | 14,924 | 10,402 | -30% | 1 | 1 | 0% | 370 | 1,861 | +403% | 0 | 0 | — |
case-11 | fail→pass | 9,081 | 11,859 | +31% | 1 | 1 | 0% | 1,453 | 2,550 | +75% | 0 | 0 | — |
case-12 | fail→pass | 9,033 | 7,331 | -19% | 1 | 1 | 0% | 1,589 | 1,329 | -16% | 0 | 0 | — |
case-13 | fail→pass | 4,131 | 10,983 | +166% | 1 | 1 | 0% | 384 | 2,421 | +530% | 0 | 0 | — |
case-14 | fail→pass | 5,174 | 11,617 | +125% | 1 | 1 | 0% | 254 | 2,192 | +763% | 0 | 0 | — |
case-15 | fail→pass | 5,302 | 6,872 | +30% | 1 | 1 | 0% | 213 | 1,585 | +644% | 0 | 0 | — |
case-16 | fail→pass | 8,180 | 8,267 | +1% | 1 | 1 | 0% | 392 | 1,553 | +296% | 0 | 0 | — |
case-17 | fail→pass | 7,169 | 8,842 | +23% | 1 | 1 | 0% | 1,234 | 1,954 | +58% | 0 | 0 | — |
case-18 | fail→pass | 9,676 | 4,912 | -49% | 1 | 1 | 0% | 662 | 1,048 | +58% | 0 | 0 | — |
case-19 | fail→pass | 5,613 | 7,001 | +25% | 1 | 1 | 0% | 915 | 1,610 | +76% | 0 | 0 | — |
case-20 | fail→pass | 5,126 | 12,523 | +144% | 1 | 1 | 0% | 220 | 2,566 | +1066% | 0 | 0 | — |
case-21 | fail→pass | 1,781 | 7,050 | +296% | 1 | 1 | 0% | 226 | 1,540 | +581% | 0 | 0 | — |
case-22 | fail→pass | 9,763 | 10,896 | +12% | 1 | 1 | 0% | 1,168 | 1,920 | +64% | 0 | 0 | — |
case-23 | fail→pass | 7,817 | 12,547 | +61% | 1 | 1 | 0% | 1,228 | 2,507 | +104% | 0 | 0 | — |
case-24 | fail→pass | 2,516 | 8,012 | +218% | 1 | 1 | 0% | 286 | 1,814 | +534% | 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. 24 cases were attempted, and 19 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +79 percentage points is the difference between those two pass rates over the 19 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.