Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →PO + Movement: generate provocations then extract useful directions (4 movement types)
.claude/skills/yogsoth-ai-provocation-and-movement/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 127% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 117% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 116% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 126% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 183% | 0% |
Generate PO provocations then extract useful directions via 4 movement types (moment-to-moment, principle extraction, focus difference, positive aspects).
| Resource | Target | Current | % | |----------|--------|---------|---| | web-search | 20 | 0 | 0% | | web-research | 5 | 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 | |--------|------| | provocation-generation | Generate PO provocations (shared) | | movement-extraction | Extract constructive directions from provocations |
| SOP | Role | |-----|------| | movement-operation | Apply 4 movement types to provocations | | stepping-stone | Use impractical provocations as stepping stones | | alternatives-generation | Generate alternatives from movement directions | | lateral-synthesis | Synthesize provocation-movement outputs |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | movement-extraction | Extract constructive directions from provocations via 4 movement types (moment-to-moment, principle, focus difference, positive aspects). |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | alternatives-generation | Generate alternatives for every known approach — ensure no approach goes unchallenged. | | lateral-synthesis | Synthesize all lateral thinking intermediate outputs into a structured idea report. | | movement-operation | Extract constructive directions from PO provocations using 4 movement types (moment-to-moment, principle, focus difference, positive aspects). | | stepping-stone | Use impractical ideas as stepping stones to reach practical solutions (de Bono Stepping Stone technique). |
<!-- 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 | fail→pass | 22,665 | 51,446 | +127% | 1 | 1 | 0% | 3,555 | 8,055 | +127% | 0 | 0 | — |
case-02 | fail→pass | 33,424 | 46,168 | +38% | 1 | 1 | 0% | 3,805 | 8,251 | +117% | 0 | 0 | — |
case-03 | pass→fail | 21,192 | 38,946 | +84% | 1 | 1 | 0% | 2,419 | 6,838 | +183% | 0 | 0 | — |
case-04 | pass→fail | 25,062 | 53,002 | +111% | 1 | 1 | 0% | 3,017 | 8,822 | +192% | 0 | 0 | — |
case-05 | pass→fail | 36,351 | 57,104 | +57% | 1 | 1 | 0% | 2,418 | 8,825 | +265% | 0 | 0 | — |
case-06 | pass→pass | 27,171 | 62,125 | +129% | 1 | 1 | 0% | 3,049 | 8,828 | +190% | 0 | 0 | — |
case-12 | fail→fail | 21,414 | 25,310 | +18% | 1 | 1 | 0% | 2,515 | 1,260 | -50% | 0 | 0 | — |
case-07 | fail→fail | 20,464 | 68,192 | +233% | 1 | 1 | 0% | 2,348 | 1,413 | -40% | 0 | 0 | — |
case-08 | pass→fail | 20,925 | 18,541 | -11% | 1 | 1 | 0% | 2,630 | 1,386 | -47% | 0 | 0 | — |
case-09 | fail→pass | 25,390 | 53,212 | +110% | 1 | 1 | 0% | 3,365 | 7,278 | +116% | 0 | 0 | — |
case-10 | pass→fail | 24,750 | 15,635 | -37% | 1 | 1 | 0% | 2,665 | 1,165 | -56% | 0 | 0 | — |
case-11 | fail→pass | 22,596 | 37,704 | +67% | 1 | 1 | 0% | 2,693 | 6,083 | +126% | 0 | 0 | — |
case-13 | pass→pass | 22,772 | 42,076 | +85% | 1 | 1 | 0% | 2,805 | 6,255 | +123% | 0 | 0 | — |
case-14 | pass→pass | 21,119 | 30,029 | +42% | 1 | 1 | 0% | 2,803 | 4,572 | +63% | 0 | 0 | — |
case-15 | pass→fail | 15,254 | 19,618 | +29% | 1 | 1 | 0% | 2,352 | 1,288 | -45% | 0 | 0 | — |
case-16 | pass→pass | 23,230 | 64,150 | +176% | 1 | 1 | 0% | 3,028 | 8,119 | +168% | 0 | 0 | — |
case-17 | pass→pass | 18,523 | 42,321 | +128% | 1 | 1 | 0% | 2,834 | 6,674 | +135% | 0 | 0 | — |
case-18 | pass→pass | 16,108 | 54,816 | +240% | 1 | 1 | 0% | 2,265 | 5,349 | +136% | 0 | 0 | — |
case-19 | pass→pass | 13,154 | 76,241 | +480% | 1 | 1 | 0% | 1,701 | 5,341 | +214% | 0 | 0 | — |
case-20 | pass→pass | 15,212 | 56,138 | +269% | 1 | 1 | 0% | 2,261 | 6,334 | +180% | 0 | 0 | — |
case-21 | pass→fail | 15,739 | 30,686 | +95% | 1 | 1 | 0% | 2,025 | 1,113 | -45% | 0 | 0 | — |
case-22 | pass→fail | 19,180 | 26,155 | +36% | 1 | 1 | 0% | 2,897 | 1,102 | -62% | 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 15 counted toward the lift figure. The other 7 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 -18 percentage points is the difference between those two pass rates over the 15 comparable cases. 10 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.