Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Ask one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.
.claude/skills/hmbown-interview/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -79% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -64% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -65% | 0% |
Use when the request lacks a material product or implementation choice that cannot be discovered from the workspace or tools.
If the user declines to answer, proceed with explicit assumptions and label them.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→pass | 12,363 | 6,109 | -51% | 1 | 1 | 0% | 2,613 | 1,355 | -48% | 0 | 0 | — |
case-01 | fail→fail | 2,823 | 5,908 | +109% | 1 | 1 | 0% | 491 | 359 | -27% | 0 | 0 | — |
case-02 | pass→pass | 5,930 | 7,245 | +22% | 1 | 1 | 0% | 1,222 | 1,422 | +16% | 0 | 0 | — |
case-03 | pass→fail | 10,654 | 3,074 | -71% | 1 | 1 | 0% | 2,599 | 333 | -87% | 0 | 0 | — |
case-04 | fail→pass | 8,110 | 2,765 | -66% | 1 | 1 | 0% | 1,661 | 626 | -62% | 0 | 0 | — |
case-05 | fail→fail | 4,847 | 6,019 | +24% | 1 | 1 | 0% | 1,020 | 1,301 | +28% | 0 | 0 | — |
case-06 | fail→pass | 13,900 | 2,900 | -79% | 1 | 1 | 0% | 3,248 | 684 | -79% | 0 | 0 | — |
case-07 | fail→pass | 9,307 | 3,154 | -66% | 1 | 1 | 0% | 1,943 | 690 | -64% | 0 | 0 | — |
case-09 | pass→pass | 5,657 | 3,426 | -39% | 1 | 1 | 0% | 1,360 | 840 | -38% | 0 | 0 | — |
case-10 | fail→pass | 12,912 | 3,685 | -71% | 1 | 1 | 0% | 2,459 | 863 | -65% | 0 | 0 | — |
case-11 | fail→fail | 4,369 | 5,189 | +19% | 1 | 1 | 0% | 582 | 1,065 | +83% | 0 | 0 | — |
case-12 | fail→pass | 4,088 | 3,021 | -26% | 1 | 1 | 0% | 670 | 671 | +0% | 0 | 0 | — |
case-13 | fail→pass | 11,902 | 2,467 | -79% | 1 | 1 | 0% | 2,788 | 574 | -79% | 0 | 0 | — |
case-14 | fail→fail | 8,222 | 7,125 | -13% | 1 | 1 | 0% | 1,686 | 1,461 | -13% | 0 | 0 | — |
case-15 | fail→fail | 1,772 | 16,258 | +817% | 1 | 1 | 0% | 286 | 3,062 | +971% | 0 | 0 | — |
case-16 | fail→pass | 21,929 | 3,006 | -86% | 1 | 1 | 0% | 4,479 | 722 | -84% | 0 | 0 | — |
case-17 | fail→fail | 8,046 | 3,629 | -55% | 1 | 1 | 0% | 1,494 | 779 | -48% | 0 | 0 | — |
case-18 | pass→pass | 9,981 | 3,040 | -70% | 1 | 1 | 0% | 1,915 | 658 | -66% | 0 | 0 | — |
case-19 | fail→pass | 10,018 | 7,253 | -28% | 1 | 1 | 0% | 2,042 | 1,548 | -24% | 0 | 0 | — |
case-20 | pass→pass | 3,880 | 3,330 | -14% | 1 | 1 | 0% | 792 | 765 | -3% | 0 | 0 | — |
case-21 | fail→fail | 11,175 | 2,957 | -74% | 1 | 1 | 0% | 2,588 | 656 | -75% | 0 | 0 | — |
case-22 | fail→fail | 11,780 | 13,657 | +16% | 1 | 1 | 0% | 663 | 1,025 | +55% | 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 21 counted toward the lift figure. The other 1 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 +36 percentage points is the difference between those two pass rates over the 21 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.