Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →DOE thinking: identify factors, define levels, and explore combinations to systematically cover the design space.
.claude/skills/yogsoth-ai-factorial-ideation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 215% | 0% |
Apply Design of Experiments (DOE) thinking to ideation: identify key factors, define their levels, and systematically explore combinations.
| Resource | Target | Current | % | |----------|--------|---------|---| | web-search | 20 | 0 | 0% | | web-research | 8 | 0 | 0% | | paper-overview | 20 | 0 | 0% | | paper-search | 12 | 0 | 0% | | paper-research | 5 | 0 | 0% |
Cannot exit strategy until ≥80% of each budget line is consumed OR yield targets are met with justification for remaining budget.
| Tactic | Role | |--------|------| | gap-driven-generation | Generate ideas for uncovered factor combinations | | evaluation-filtering | Score and filter factorial-derived ideas |
| SOP | Role | |-----|------| | factor-level-design | Identify factors, define levels, build experiment matrix | | coverage-gap-detection | Find unexplored regions in factorial space | | failure-driven-generation | Generate solutions for problematic combinations | | enumeration-synthesis | Synthesize factorial exploration into idea report |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | gap-driven-generation | Generate solutions targeting specific coverage gaps — detect gaps, generate failure-driven solutions, and design factor-level experiments. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | coverage-gap-detection | Detect uncovered regions in the solution space, producing a prioritized gap list. | | creative-ideation-factor-level-design | Identify factors and their levels for a problem, then design an experiment matrix for systematic exploration. | | enumeration-synthesis | Synthesize all systematic enumeration outputs into a structured idea report with prioritized recommendations. |
<!-- END available-tables (generated) -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→fail | 9,791 | 18,360 | +88% | 1 | 1 | 0% | 1,636 | 2,853 | +74% | 0 | 0 | — |
case-02 | pass→fail | 16,891 | 33,546 | +99% | 1 | 1 | 0% | 3,606 | 7,262 | +101% | 0 | 0 | — |
case-03 | pass→pass | 14,832 | 23,261 | +57% | 1 | 1 | 0% | 2,139 | 4,160 | +94% | 0 | 0 | — |
case-04 | fail→pass | 11,777 | 6,710 | -43% | 1 | 1 | 0% | 1,795 | 1,660 | -8% | 0 | 0 | — |
case-05 | pass→pass | 9,205 | 6,647 | -28% | 1 | 1 | 0% | 1,327 | 1,614 | +22% | 0 | 0 | — |
case-06 | pass→pass | 7,480 | 8,214 | +10% | 1 | 1 | 0% | 1,209 | 1,984 | +64% | 0 | 0 | — |
case-07 | fail→fail | 12,338 | 3,744 | -70% | 1 | 1 | 0% | 2,041 | 1,256 | -38% | 0 | 0 | — |
case-08 | fail→fail | 18,637 | 9,366 | -50% | 1 | 1 | 0% | 2,769 | 1,151 | -58% | 0 | 0 | — |
case-09 | pass→pass | 11,580 | 6,100 | -47% | 1 | 1 | 0% | 1,741 | 1,716 | -1% | 0 | 0 | — |
case-10 | pass→fail | 11,992 | 5,357 | -55% | 1 | 1 | 0% | 1,827 | 1,500 | -18% | 0 | 0 | — |
case-11 | pass→pass | 10,380 | 3,424 | -67% | 1 | 1 | 0% | 1,618 | 1,102 | -32% | 0 | 0 | — |
case-12 | fail→pass | 17,151 | 2,142 | -88% | 1 | 1 | 0% | 1,289 | 979 | -24% | 0 | 0 | — |
case-13 | fail→pass | 5,204 | 2,818 | -46% | 1 | 1 | 0% | 733 | 1,008 | +38% | 0 | 0 | — |
case-14 | fail→pass | 15,889 | 1,620 | -90% | 1 | 1 | 0% | 811 | 906 | +12% | 0 | 0 | — |
case-15 | fail→pass | 2,842 | 2,992 | +5% | 1 | 1 | 0% | 354 | 1,116 | +215% | 0 | 0 | — |
case-16 | fail→pass | 4,124 | 1,848 | -55% | 1 | 1 | 0% | 596 | 935 | +57% | 0 | 0 | — |
case-17 | fail→pass | 21,780 | 3,109 | -86% | 1 | 1 | 0% | 875 | 1,026 | +17% | 0 | 0 | — |
case-18 | pass→pass | 9,086 | 2,108 | -77% | 1 | 1 | 0% | 1,278 | 914 | -28% | 0 | 0 | — |
case-19 | fail→pass | 12,180 | 9,131 | -25% | 1 | 1 | 0% | 1,908 | 1,981 | +4% | 0 | 0 | — |
case-20 | fail→fail | 7,987 | 3,631 | -55% | 1 | 1 | 0% | 1,306 | 1,212 | -7% | 0 | 0 | — |
case-21 | fail→fail | 8,286 | 4,227 | -49% | 1 | 1 | 0% | 1,286 | 1,188 | -8% | 0 | 0 | — |
case-22 | pass→pass | 10,675 | 4,952 | -54% | 1 | 1 | 0% | 1,505 | 1,371 | -9% | 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 19 counted toward the lift figure. The other 3 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 19 comparable cases. 3 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.