Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when reviewing CI coverage, automated checks, or test strategy related to Follow mocking best practices. Focus on whether the rule is continuously verified, not just documented.
.claude/skills/thedaviddias-mock-best-practices/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 26% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -21% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 2% | 0% |
Proper mocking isolates units under test while keeping tests realistic—over-mocking creates false confidence when tests pass but production breaks.
Review this test file for mocking patterns, checking for over-mocking, missing mock cleanup, and proper mock implementation.
Improve mocking strategy by removing unnecessary mocks, adding proper cleanup, and mocking at the right level of abstraction.
Explain mocking best practices including when to mock, what to mock, and common mocking pitfalls.
Review tests, CI workflows, and enforcement points related to Follow mocking best practices. Flag exact gaps where the rule is not automatically verified or where failures do not block regressions.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/testing/mock-best-practices
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | fail→pass | 11,358 | 9,293 | -18% | 1 | 1 | 0% | 2,025 | 1,868 | -8% | 0 | 0 | — |
case-01 | fail→pass | 11,049 | 12,283 | +11% | 1 | 1 | 0% | 2,305 | 2,711 | +18% | 0 | 0 | — |
case-02 | pass→pass | 4,186 | 3,672 | -12% | 1 | 1 | 0% | 776 | 974 | +26% | 0 | 0 | — |
case-03 | pass→pass | 12,661 | 8,819 | -30% | 1 | 1 | 0% | 2,449 | 1,935 | -21% | 0 | 0 | — |
case-04 | pass→pass | 13,331 | 10,889 | -18% | 1 | 1 | 0% | 2,239 | 2,286 | +2% | 0 | 0 | — |
case-05 | pass→pass | 9,333 | 7,612 | -18% | 1 | 1 | 0% | 1,802 | 1,635 | -9% | 0 | 0 | — |
case-07 | pass→pass | 10,494 | 7,841 | -25% | 1 | 1 | 0% | 1,928 | 1,728 | -10% | 0 | 0 | — |
case-08 | pass→pass | 5,170 | 5,924 | +15% | 1 | 1 | 0% | 896 | 1,310 | +46% | 0 | 0 | — |
case-09 | pass→pass | 10,845 | 8,660 | -20% | 1 | 1 | 0% | 2,055 | 1,941 | -6% | 0 | 0 | — |
case-10 | pass→pass | 9,987 | 5,999 | -40% | 1 | 1 | 0% | 1,653 | 1,204 | -27% | 0 | 0 | — |
case-15 | pass→pass | 10,362 | 7,618 | -26% | 1 | 1 | 0% | 1,943 | 1,611 | -17% | 0 | 0 | — |
case-11 | pass→pass | 10,366 | 9,349 | -10% | 1 | 1 | 0% | 1,709 | 1,775 | +4% | 0 | 0 | — |
case-12 | pass→pass | 6,027 | 4,554 | -24% | 1 | 1 | 0% | 1,113 | 1,036 | -7% | 0 | 0 | — |
case-13 | pass→pass | 10,382 | 12,374 | +19% | 1 | 1 | 0% | 1,850 | 2,866 | +55% | 0 | 0 | — |
case-14 | pass→pass | 12,690 | 10,185 | -20% | 1 | 1 | 0% | 2,434 | 2,452 | +1% | 0 | 0 | — |
case-16 | pass→pass | 12,669 | 12,503 | -1% | 1 | 1 | 0% | 2,107 | 2,433 | +15% | 0 | 0 | — |
case-17 | pass→pass | 11,573 | 10,162 | -12% | 1 | 1 | 0% | 2,324 | 2,308 | -1% | 0 | 0 | — |
case-18 | pass→pass | 8,570 | 7,045 | -18% | 1 | 1 | 0% | 1,635 | 1,501 | -8% | 0 | 0 | — |
case-19 | pass→pass | 13,820 | 14,804 | +7% | 1 | 1 | 0% | 2,364 | 2,824 | +19% | 0 | 0 | — |
case-20 | pass→pass | 4,737 | 5,167 | +9% | 1 | 1 | 0% | 871 | 1,219 | +40% | 0 | 0 | — |
case-21 | pass→pass | 12,068 | 8,152 | -32% | 1 | 1 | 0% | 2,305 | 1,761 | -24% | 0 | 0 | — |
case-22 | pass→pass | 7,944 | 6,952 | -12% | 1 | 1 | 0% | 1,479 | 1,570 | +6% | 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 +9 percentage points is the difference between those two pass rates over the 22 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.