Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Identify technical and physical contradictions in a system through functional modeling and matrix analysis.
.claude/skills/yogsoth-ai-contradiction-identification/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 38% | 0% |
Identify technical and physical contradictions in a system, providing resolution paths for each.
Build substance-field model of the system using function-model-construction SOP. Annotate useful, harmful, and insufficient interactions.
For each identified technical contradiction (improving parameter vs worsening parameter), query the 39×39 TRIZ contradiction matrix via contradiction-matrix-lookup SOP.
For physical contradictions (same parameter must be both X and not-X), identify applicable separation principles (time, space, condition, scale) via separation-principle SOP.
| Metric | Floor | |--------|-------| | Technical contradictions identified | ≥2 | | Physical contradictions identified | ≥1 | | Resolution paths per contradiction | ≥2 | | Applicable principles listed | ≥4 |
| SOP | Role | |-----|------| | function-model-construction | Stage 1 — build substance-field model | | contradiction-matrix-lookup | Stage 2 — query matrix for principles | | separation-principle | Stage 3 — resolve physical contradictions | | triz-principle-application | Post — apply selected principles |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | contradiction-matrix-lookup | Query the 39x39 TRIZ contradiction matrix to find recommended inventive principles for a given technical contradiction. | | function-model-construction | Build substance-field functional model of a system, annotating useful, harmful, insufficient, and excessive interactions. | | separation-principle | Apply time/space/condition/scale separation to resolve physical contradictions where the same parameter must satisfy opposing requirements. |
<!-- 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 | 22,152 | 29,785 | +34% | 1 | 1 | 0% | 3,229 | 4,988 | +54% | 0 | 0 | — |
case-02 | fail→pass | 21,568 | 29,839 | +38% | 1 | 1 | 0% | 3,446 | 5,084 | +48% | 0 | 0 | — |
case-03 | fail→pass | 23,343 | 26,526 | +14% | 1 | 1 | 0% | 3,687 | 4,840 | +31% | 0 | 0 | — |
case-04 | pass→pass | 14,354 | 26,113 | +82% | 1 | 1 | 0% | 2,126 | 4,604 | +117% | 0 | 0 | — |
case-05 | pass→fail | 13,637 | 36,082 | +165% | 1 | 1 | 0% | 2,597 | 6,619 | +155% | 0 | 0 | — |
case-06 | pass→fail | 15,519 | 27,769 | +79% | 1 | 1 | 0% | 2,893 | 5,349 | +85% | 0 | 0 | — |
case-07 | fail→pass | 29,481 | 34,463 | +17% | 1 | 1 | 0% | 4,357 | 6,200 | +42% | 0 | 0 | — |
case-08 | fail→pass | 20,152 | 26,189 | +30% | 1 | 1 | 0% | 3,089 | 4,275 | +38% | 0 | 0 | — |
case-09 | fail→pass | 19,718 | 25,814 | +31% | 1 | 1 | 0% | 2,982 | 4,689 | +57% | 0 | 0 | — |
case-10 | fail→pass | 21,573 | 22,588 | +5% | 1 | 1 | 0% | 3,520 | 3,870 | +10% | 0 | 0 | — |
case-11 | fail→pass | 14,066 | 27,150 | +93% | 1 | 1 | 0% | 2,315 | 4,311 | +86% | 0 | 0 | — |
case-12 | fail→pass | 20,766 | 35,929 | +73% | 1 | 1 | 0% | 3,409 | 6,521 | +91% | 0 | 0 | — |
case-13 | fail→pass | 15,728 | 22,051 | +40% | 1 | 1 | 0% | 2,306 | 4,181 | +81% | 0 | 0 | — |
case-14 | fail→pass | 22,321 | 31,621 | +42% | 1 | 1 | 0% | 3,685 | 5,698 | +55% | 0 | 0 | — |
case-15 | fail→pass | 22,420 | 25,755 | +15% | 1 | 1 | 0% | 3,302 | 4,534 | +37% | 0 | 0 | — |
case-16 | fail→pass | 17,518 | 24,833 | +42% | 1 | 1 | 0% | 2,787 | 4,661 | +67% | 0 | 0 | — |
case-17 | fail→pass | 13,283 | 22,013 | +66% | 1 | 1 | 0% | 2,212 | 4,086 | +85% | 0 | 0 | — |
case-18 | fail→pass | 27,080 | 27,555 | +2% | 1 | 1 | 0% | 4,171 | 4,516 | +8% | 0 | 0 | — |
case-19 | fail→pass | 21,809 | 25,645 | +18% | 1 | 1 | 0% | 3,414 | 4,515 | +32% | 0 | 0 | — |
case-20 | fail→pass | 20,531 | 21,541 | +5% | 1 | 1 | 0% | 3,028 | 3,935 | +30% | 0 | 0 | — |
case-21 | fail→pass | 20,299 | 24,043 | +18% | 1 | 1 | 0% | 3,098 | 4,381 | +41% | 0 | 0 | — |
case-22 | fail→pass | 21,571 | 22,434 | +4% | 1 | 1 | 0% | 2,959 | 4,127 | +39% | 0 | 0 | — |
case-23 | fail→pass | 18,916 | 28,793 | +52% | 1 | 1 | 0% | 2,797 | 5,157 | +84% | 0 | 0 | — |
case-24 | fail→pass | 18,670 | 23,530 | +26% | 1 | 1 | 0% | 2,874 | 4,033 | +40% | 0 | 0 | — |
case-25 | fail→pass | 18,147 | 23,710 | +31% | 1 | 1 | 0% | 2,897 | 4,421 | +53% | 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. 25 cases were attempted. The headline lift of +80 percentage points is the difference between those two pass rates over the 25 comparable cases. 2 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.