Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Hegelian thesis-antithesis-synthesis cycle — propose position, generate opposition, structured debate, synthesize transcending insight. Combines evaporating-cloud and polarity-mapping SOPs.
.claude/skills/yogsoth-ai-dialectical-synthesis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 58% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 51% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 106% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 25% | 0% |
Thesis-antithesis-synthesis cycle for tension resolution.
Subagent: evaporating-cloud, polarity-mapping Import: paper-search, web-research
Model conflict as evaporating cloud (expose hidden assumptions behind opposing needs). If tension is a genuine polarity (not resolvable), map it with polarity management framework. Distinguish solvable conflicts from inherent polarities.
<HARD-GATE>
- evaporating clouds: >= 1
- assumptions exposed: >= 3
- polarity maps (if applicable): >= 1
</HARD-GATE><!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | deep-insight-paper-search | AI-powered paper summary and search. Import of literature-engine/literature-search skill. AI summary level — cite as "AI-extracted" not "paper states". | | deep-insight-web-research | Deep web research with full page fetching via Apify. Import of web-browsing/web-research skill. Must fetch full page — no conclusions from previews. | | evaporating-cloud | Model conflicts as Goldratt's Evaporating Cloud — expose hidden assumptions behind opposing needs to dissolve the conflict. | | polarity-mapping | Map unresolvable tensions as Johnson polarities — 4 quadrants (positive/negative of each pole), early warnings, action steps for managing rather than solving. |
<!-- 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-04 | pass→pass | 11,975 | 24,716 | +106% | 1 | 1 | 0% | 2,204 | 4,498 | +104% | 0 | 0 | — |
case-01 | fail→pass | 19,693 | 25,685 | +30% | 1 | 1 | 0% | 2,800 | 4,412 | +58% | 0 | 0 | — |
case-02 | fail→pass | 18,277 | 24,796 | +36% | 1 | 1 | 0% | 2,821 | 4,270 | +51% | 0 | 0 | — |
case-03 | fail→pass | 22,118 | 27,157 | +23% | 1 | 1 | 0% | 3,414 | 4,445 | +30% | 0 | 0 | — |
case-05 | pass→pass | 10,468 | 20,422 | +95% | 1 | 1 | 0% | 1,895 | 3,687 | +95% | 0 | 0 | — |
case-06 | pass→fail | 11,831 | 24,852 | +110% | 1 | 1 | 0% | 2,004 | 4,538 | +126% | 0 | 0 | — |
case-07 | fail→pass | 18,085 | 33,269 | +84% | 1 | 1 | 0% | 2,678 | 5,508 | +106% | 0 | 0 | — |
case-08 | fail→pass | 23,708 | 24,519 | +3% | 1 | 1 | 0% | 3,494 | 4,354 | +25% | 0 | 0 | — |
case-09 | pass→pass | 20,689 | 29,315 | +42% | 1 | 1 | 0% | 3,292 | 5,019 | +52% | 0 | 0 | — |
case-10 | fail→pass | 21,886 | 32,492 | +48% | 1 | 1 | 0% | 3,043 | 5,544 | +82% | 0 | 0 | — |
case-11 | fail→pass | 16,660 | 24,388 | +46% | 1 | 1 | 0% | 2,493 | 4,076 | +63% | 0 | 0 | — |
case-12 | fail→pass | 21,452 | 32,580 | +52% | 1 | 1 | 0% | 3,371 | 5,537 | +64% | 0 | 0 | — |
case-13 | fail→pass | 17,941 | 38,276 | +113% | 1 | 1 | 0% | 2,839 | 6,568 | +131% | 0 | 0 | — |
case-14 | fail→pass | 16,525 | 25,142 | +52% | 1 | 1 | 0% | 2,551 | 4,142 | +62% | 0 | 0 | — |
case-15 | fail→pass | 19,758 | 24,319 | +23% | 1 | 1 | 0% | 2,944 | 4,173 | +42% | 0 | 0 | — |
case-16 | fail→pass | 22,620 | 20,646 | -9% | 1 | 1 | 0% | 3,524 | 3,435 | -3% | 0 | 0 | — |
case-17 | fail→pass | 22,370 | 24,500 | +10% | 1 | 1 | 0% | 3,503 | 4,251 | +21% | 0 | 0 | — |
case-18 | fail→pass | 18,482 | 22,332 | +21% | 1 | 1 | 0% | 2,714 | 3,784 | +39% | 0 | 0 | — |
case-19 | fail→pass | 15,832 | 21,739 | +37% | 1 | 1 | 0% | 2,224 | 3,910 | +76% | 0 | 0 | — |
case-20 | fail→pass | 19,046 | 27,439 | +44% | 1 | 1 | 0% | 2,856 | 4,746 | +66% | 0 | 0 | — |
case-21 | fail→pass | 16,142 | 34,086 | +111% | 1 | 1 | 0% | 2,446 | 5,662 | +131% | 0 | 0 | — |
case-22 | fail→pass | 24,361 | 25,631 | +5% | 1 | 1 | 0% | 3,766 | 4,411 | +17% | 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 +77 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.