Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate PO provocations and extract constructive movement. Orchestrates assumption surfacing → provocation creation → movement extraction → idea formation.
.claude/skills/yogsoth-ai-creative-ideation-provocation-generation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 165% | 0% |
Generate PO provocations and extract constructive movement.
Identify the dominant assumptions and patterns in the current thinking using assumption-surfacing SOP. Prioritize paradigmatic assumptions with high disruption potential.
Apply PO operators (escape, reversal, exaggeration, distortion, wishful thinking) to the surfaced assumptions using po-provocation SOP. Generate ≥10 provocations across all 5 types.
For each promising provocation, extract constructive movement using movement-operation SOP (if available) or direct reasoning. Four movement types:
Convert movement directions into concrete ideas using idea-synthesis SOP. Each idea must have a clear mechanism, not just a vague direction.
| Metric | Floor | |--------|-------| | Assumptions surfaced | ≥5 | | Provocations generated | ≥10 | | Movement directions extracted | ≥6 | | Concrete ideas formed | ≥4 |
| SOP | Role | |-----|------| | assumption-surfacing | Stage 1 — enumerate assumptions | | po-provocation | Stage 2 — generate PO statements | | random-word-stimulus | Stage 2 — alternative stimulus source | | movement-operation | Stage 3 — extract movement (if available) | | idea-synthesis | Stage 4 — form concrete ideas | | novelty-scoring | Post — score resulting ideas | | saturation-detection | Post — check if more provocations needed |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-12 | pass→pass | 16,550 | 11,989 | -28% | 1 | 1 | 0% | 2,742 | 2,419 | -12% | 0 | 0 | — |
case-01 | fail→pass | 26,686 | 22,033 | -17% | 1 | 1 | 0% | 4,177 | 4,080 | -2% | 0 | 0 | — |
case-02 | fail→pass | 23,899 | 21,800 | -9% | 1 | 1 | 0% | 3,844 | 4,274 | +11% | 0 | 0 | — |
case-03 | fail→pass | 25,052 | 26,789 | +7% | 1 | 1 | 0% | 3,877 | 4,933 | +27% | 0 | 0 | — |
case-04 | fail→fail | 9,361 | 11,480 | +23% | 1 | 1 | 0% | 1,415 | 2,198 | +55% | 0 | 0 | — |
case-05 | fail→fail | 11,766 | 20,088 | +71% | 1 | 1 | 0% | 1,869 | 3,766 | +101% | 0 | 0 | — |
case-13 | fail→pass | 16,146 | 21,582 | +34% | 1 | 1 | 0% | 2,297 | 3,692 | +61% | 0 | 0 | — |
case-06 | fail→fail | 3,512 | 6,479 | +84% | 1 | 1 | 0% | 561 | 1,365 | +143% | 0 | 0 | — |
case-07 | fail→pass | 9,217 | 23,119 | +151% | 1 | 1 | 0% | 1,537 | 4,077 | +165% | 0 | 0 | — |
case-08 | fail→pass | 16,336 | 18,907 | +16% | 1 | 1 | 0% | 2,511 | 3,333 | +33% | 0 | 0 | — |
case-09 | pass→fail | 14,634 | 24,360 | +66% | 1 | 1 | 0% | 2,458 | 4,461 | +81% | 0 | 0 | — |
case-10 | pass→fail | 16,956 | 42,552 | +151% | 1 | 1 | 0% | 3,286 | 5,903 | +80% | 0 | 0 | — |
case-11 | fail→fail | 6,254 | 4,934 | -21% | 1 | 1 | 0% | 986 | 1,119 | +13% | 0 | 0 | — |
case-14 | pass→pass | 15,212 | 21,478 | +41% | 1 | 1 | 0% | 2,286 | 3,433 | +50% | 0 | 0 | — |
case-15 | fail→fail | 10,951 | 11,692 | +7% | 1 | 1 | 0% | 1,599 | 1,991 | +25% | 0 | 0 | — |
case-16 | pass→pass | 12,617 | 20,711 | +64% | 1 | 1 | 0% | 1,839 | 3,130 | +70% | 0 | 0 | — |
case-17 | pass→pass | 22,097 | 18,229 | -18% | 1 | 1 | 0% | 3,085 | 3,293 | +7% | 0 | 0 | — |
case-18 | pass→pass | 39,622 | 33,272 | -16% | 1 | 1 | 0% | 3,174 | 5,515 | +74% | 0 | 0 | — |
case-19 | fail→pass | 19,955 | 32,623 | +63% | 1 | 1 | 0% | 3,156 | 5,701 | +81% | 0 | 0 | — |
case-20 | pass→pass | 14,849 | 18,679 | +26% | 1 | 1 | 0% | 2,309 | 3,385 | +47% | 0 | 0 | — |
case-21 | fail→fail | 5,828 | 6,383 | +10% | 1 | 1 | 0% | 925 | 1,342 | +45% | 0 | 0 | — |
case-22 | pass→fail | 13,959 | 8,770 | -37% | 1 | 1 | 0% | 2,440 | 1,795 | -26% | 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 +18 percentage points is the difference between those two pass rates over the 22 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.