Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Test quality review drawing on twelve classic engineering books — with primary focus on xUnit Test Patterns, The Art of Unit Testing, How Google Tests Software, and Working Effectively with Legacy Code — that diagnoses structural problems in an existing test suite: brittleness, mock...
.claude/skills/sickn33-brooks-test/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 208% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 218% | 0% |
| case-08 | ✓→✗ | ▼ Worse | -72% | 0% |
| case-12 | ✓→✗ | ▼ Worse | -83% | 0% |
| case-13 | ✓→✗ | ▼ Worse | -80% | 0% |
Use this skill when you need test quality review drawing on twelve classic engineering books — with primary focus on xUnit Test Patterns, The Art of Unit Testing, How Google Tests Software, and Working Effectively with Legacy Code — that diagnoses structural problems in an existing test suite: brittleness, mock...
../_shared/common.md for the Iron Law, Project Config, Report Template, and Health Score rules../_shared/source-coverage.md for book-level coverage, exceptions, and tradeoffs../_shared/test-decay-risks.md for test-space symptom definitions and source attributionstest-guide.md in this directory for the test quality review frameworkIf the user has not shared test files or pointed to a test directory: apply Auto Scope Detection from ../_shared/common.md to determine the review scope before proceeding.
Mode line in report: Test Quality Review
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,478 | 4,183 | -82% | 1 | 1 | 0% | 4,451 | 630 | -86% | 0 | 0 | — |
case-02 | fail→pass | 9,894 | 27,688 | +180% | 1 | 1 | 0% | 1,773 | 5,467 | +208% | 0 | 0 | — |
case-03 | fail→fail | 6,227 | 3,753 | -40% | 1 | 1 | 0% | 914 | 553 | -39% | 0 | 0 | — |
case-04 | pass→pass | 12,545 | 15,126 | +21% | 1 | 1 | 0% | 2,901 | 3,481 | +20% | 0 | 0 | — |
case-05 | pass→pass | 11,659 | 15,363 | +32% | 1 | 1 | 0% | 2,417 | 3,379 | +40% | 0 | 0 | — |
case-06 | fail→pass | 5,580 | 14,534 | +160% | 1 | 1 | 0% | 1,146 | 3,647 | +218% | 0 | 0 | — |
case-07 | fail→fail | 23,037 | 3,824 | -83% | 1 | 1 | 0% | 3,550 | 539 | -85% | 0 | 0 | — |
case-08 | pass→fail | 14,089 | 5,075 | -64% | 1 | 1 | 0% | 2,357 | 665 | -72% | 0 | 0 | — |
case-09 | fail→fail | 9,406 | 3,788 | -60% | 1 | 1 | 0% | 1,581 | 623 | -61% | 0 | 0 | — |
case-10 | fail→fail | 18,273 | 4,890 | -73% | 1 | 1 | 0% | 3,372 | 588 | -83% | 0 | 0 | — |
case-11 | fail→fail | 10,097 | 12,712 | +26% | 1 | 1 | 0% | 1,778 | 2,684 | +51% | 0 | 0 | — |
case-12 | pass→fail | 16,003 | 3,744 | -77% | 1 | 1 | 0% | 3,148 | 536 | -83% | 0 | 0 | — |
case-13 | pass→fail | 16,316 | 3,746 | -77% | 1 | 1 | 0% | 2,831 | 574 | -80% | 0 | 0 | — |
case-14 | fail→fail | 8,802 | 7,946 | -10% | 1 | 1 | 0% | 1,709 | 2,001 | +17% | 0 | 0 | — |
case-15 | pass→fail | 15,915 | 4,851 | -70% | 1 | 1 | 0% | 2,789 | 736 | -74% | 0 | 0 | — |
case-16 | pass→fail | 19,960 | 7,851 | -61% | 1 | 1 | 0% | 3,562 | 598 | -83% | 0 | 0 | — |
case-17 | pass→fail | 9,779 | 8,971 | -8% | 1 | 1 | 0% | 2,097 | 632 | -70% | 0 | 0 | — |
case-18 | fail→fail | 8,471 | 4,655 | -45% | 1 | 1 | 0% | 1,282 | 623 | -51% | 0 | 0 | — |
case-19 | fail→fail | 7,777 | 5,459 | -30% | 1 | 1 | 0% | 1,194 | 629 | -47% | 0 | 0 | — |
case-20 | pass→pass | 17,434 | 35,550 | +104% | 1 | 1 | 0% | 3,048 | 6,530 | +114% | 0 | 0 | — |
case-21 | fail→fail | 11,144 | 12,931 | +16% | 1 | 1 | 0% | 1,021 | 2,211 | +117% | 0 | 0 | — |
case-22 | pass→fail | 13,413 | 4,099 | -69% | 1 | 1 | 0% | 2,540 | 580 | -77% | 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, and 8 counted toward the lift figure. The other 14 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 -23 percentage points is the difference between those two pass rates over the 8 comparable cases. 8 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.