Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply when facing a novel UI interaction or architectural decision with no precedent in the codebase. Build 2-3 competing prototypes and compare side by side before committing.
.claude/skills/kunanonj-cursor-plugin-pstack-principle-exhaust-the-design-space/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 15% | 0% |
When a novel interaction or architectural decision has no established precedent, explore several concrete alternatives before implementation. Building the wrong thing costs more than exploring three options.
The rule: When the right answer is not obvious, build 2-3 competing prototypes or sketches. Compare them side by side. Only then commit.
When it applies:
When it doesn't:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 7,562 | 3,304 | -56% | 1 | 1 | 0% | 1,466 | 779 | -47% | 0 | 0 | — |
case-01 | fail→pass | 16,410 | 19,890 | +21% | 1 | 1 | 0% | 3,933 | 3,755 | -5% | 0 | 0 | — |
case-02 | fail→pass | 10,880 | 12,315 | +13% | 1 | 1 | 0% | 1,724 | 2,371 | +38% | 0 | 0 | — |
case-03 | fail→pass | 11,656 | 11,123 | -5% | 1 | 1 | 0% | 2,070 | 2,404 | +16% | 0 | 0 | — |
case-04 | fail→pass | 9,109 | 13,135 | +44% | 1 | 1 | 0% | 1,730 | 2,390 | +38% | 0 | 0 | — |
case-05 | fail→pass | 11,893 | 15,455 | +30% | 1 | 1 | 0% | 2,669 | 3,067 | +15% | 0 | 0 | — |
case-06 | fail→pass | 16,635 | 16,861 | +1% | 1 | 1 | 0% | 3,152 | 3,789 | +20% | 0 | 0 | — |
case-07 | fail→pass | 28,318 | 28,871 | +2% | 1 | 1 | 0% | 6,180 | 6,227 | +1% | 0 | 0 | — |
case-08 | fail→pass | 7,100 | 10,699 | +51% | 1 | 1 | 0% | 1,303 | 1,864 | +43% | 0 | 0 | — |
case-09 | fail→pass | 13,672 | 18,103 | +32% | 1 | 1 | 0% | 2,812 | 3,886 | +38% | 0 | 0 | — |
case-10 | fail→pass | 19,202 | 21,859 | +14% | 1 | 1 | 0% | 4,381 | 4,595 | +5% | 0 | 0 | — |
case-11 | fail→pass | 17,523 | 16,309 | -7% | 1 | 1 | 0% | 4,081 | 3,763 | -8% | 0 | 0 | — |
case-12 | fail→pass | 12,988 | 12,375 | -5% | 1 | 1 | 0% | 2,573 | 2,414 | -6% | 0 | 0 | — |
case-13 | fail→pass | 11,148 | 11,533 | +3% | 1 | 1 | 0% | 2,110 | 2,297 | +9% | 0 | 0 | — |
case-14 | fail→pass | 20,782 | 14,876 | -28% | 1 | 1 | 0% | 4,466 | 3,077 | -31% | 0 | 0 | — |
case-15 | fail→pass | 19,266 | 12,074 | -37% | 1 | 1 | 0% | 4,167 | 2,387 | -43% | 0 | 0 | — |
case-16 | fail→pass | 20,511 | 25,048 | +22% | 1 | 1 | 0% | 4,676 | 5,198 | +11% | 0 | 0 | — |
case-17 | fail→pass | 13,902 | 13,838 | -0% | 1 | 1 | 0% | 2,428 | 2,539 | +5% | 0 | 0 | — |
case-18 | fail→pass | 5,138 | 14,425 | +181% | 1 | 1 | 0% | 952 | 2,718 | +186% | 0 | 0 | — |
case-19 | fail→pass | 28,623 | 19,463 | -32% | 1 | 1 | 0% | 6,176 | 4,168 | -33% | 0 | 0 | — |
case-21 | pass→pass | 1,190 | 1,754 | +47% | 1 | 1 | 0% | 173 | 390 | +125% | 0 | 0 | — |
case-22 | fail→fail | 6,102 | 7,086 | +16% | 1 | 1 | 0% | 1,236 | 1,359 | +10% | 0 | 0 | — |
case-23 | pass→pass | 8,704 | 4,869 | -44% | 1 | 1 | 0% | 1,958 | 1,257 | -36% | 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 +83 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.