Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Reverse statements → extract insights → build constructive alternatives. Systematic inversion pipeline from negation to innovation.
.claude/skills/yogsoth-ai-inversion-protocol/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 110% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 63% | 0% |
Reverse statements → extract insights → build constructive alternatives.
Use reversal-generation SOP to systematically reverse positive statements, assumptions, and goals into their opposites. Generate both simple negations and creative inversions.
For each reversal, use inversion-extraction SOP to extract constructive insights. Ask: "What does this reversal reveal about the original? What hidden constraint does it expose?"
Use constructive-rebellion SOP to build viable, constructive alternatives from the extracted insights. Transform destructive negation into productive innovation.
| Metric | Floor | |--------|-------| | Reversals generated | ≥6 | | Insights extracted | ≥5 | | Constructive alternatives | ≥4 | | Alternatives with clear mechanism | ≥3 |
| SOP | Role | |-----|------| | reversal-generation | Stage 1 — systematic reversal of statements | | inversion-extraction | Stage 2 — extract insights from reversals | | constructive-rebellion | Stage 3 — build alternatives from insights | | worst-case-design | Stage 1 alt — design worst solution for inversion | | destruction-synthesis | Post — synthesize all inversion outputs |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | constructive-rebellion | Build constructive alternatives from destructive negation. Transform violated assumptions into viable innovation directions. | | inversion-extraction | Extract constructive insights from worst solutions. Transform failure analysis into innovation directions. | | reversal-generation | Systematically reverse positive statements to generate creative inversions. Produces reversed statements with initial associations. | | worst-case-design | Design the worst possible solution. Deliberate failure engineering to reveal hidden constraints and inversion opportunities. |
<!-- 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 | 27,593 | 28,742 | +4% | 1 | 1 | 0% | 3,443 | 3,936 | +14% | 0 | 0 | — |
case-02 | fail→pass | 32,922 | 36,239 | +10% | 1 | 1 | 0% | 4,134 | 5,630 | +36% | 0 | 0 | — |
case-03 | fail→pass | 47,436 | 37,917 | -20% | 1 | 1 | 0% | 2,999 | 3,862 | +29% | 0 | 0 | — |
case-04 | pass→fail | 23,875 | 30,343 | +27% | 1 | 1 | 0% | 1,242 | 4,648 | +274% | 0 | 0 | — |
case-05 | pass→fail | 53,664 | 32,788 | -39% | 1 | 1 | 0% | 2,152 | 4,276 | +99% | 0 | 0 | — |
case-06 | pass→fail | 19,468 | 32,850 | +69% | 1 | 1 | 0% | 2,276 | 4,427 | +95% | 0 | 0 | — |
case-07 | fail→pass | 19,714 | 19,409 | -2% | 1 | 1 | 0% | 1,346 | 2,822 | +110% | 0 | 0 | — |
case-08 | fail→pass | 35,471 | 35,097 | -1% | 1 | 1 | 0% | 2,229 | 3,643 | +63% | 0 | 0 | — |
case-09 | fail→pass | 33,436 | 32,090 | -4% | 1 | 1 | 0% | 1,488 | 2,875 | +93% | 0 | 0 | — |
case-10 | pass→pass | 28,000 | 26,391 | -6% | 1 | 1 | 0% | 1,811 | 2,814 | +55% | 0 | 0 | — |
case-11 | pass→pass | 34,882 | 29,426 | -16% | 1 | 1 | 0% | 3,090 | 4,180 | +35% | 0 | 0 | — |
case-12 | pass→pass | 37,854 | 43,933 | +16% | 1 | 1 | 0% | 3,499 | 4,372 | +25% | 0 | 0 | — |
case-13 | fail→pass | 38,478 | 70,129 | +82% | 1 | 1 | 0% | 2,817 | 4,222 | +50% | 0 | 0 | — |
case-14 | fail→pass | 38,725 | 17,843 | -54% | 1 | 1 | 0% | 2,807 | 1,043 | -63% | 0 | 0 | — |
case-15 | fail→pass | 44,441 | 49,003 | +10% | 1 | 1 | 0% | 2,070 | 1,040 | -50% | 0 | 0 | — |
case-16 | fail→pass | 17,715 | 4,345 | -75% | 1 | 1 | 0% | 1,861 | 1,112 | -40% | 0 | 0 | — |
case-17 | pass→pass | 34,998 | 32,956 | -6% | 1 | 1 | 0% | 4,433 | 4,857 | +10% | 0 | 0 | — |
case-18 | pass→pass | 18,797 | 12,944 | -31% | 1 | 1 | 0% | 2,271 | 2,030 | -11% | 0 | 0 | — |
case-19 | pass→pass | 42,023 | 27,325 | -35% | 1 | 1 | 0% | 6,875 | 4,043 | -41% | 0 | 0 | — |
case-20 | pass→pass | 21,840 | 28,957 | +33% | 1 | 1 | 0% | 2,704 | 4,409 | +63% | 0 | 0 | — |
case-21 | pass→pass | 26,723 | 40,156 | +50% | 1 | 1 | 0% | 2,844 | 6,018 | +112% | 0 | 0 | — |
case-22 | pass→pass | 25,533 | 30,932 | +21% | 1 | 1 | 0% | 2,868 | 4,503 | +57% | 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 +32 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.