Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Surface Argyris governing variables and test whether the problem dissolves under alternative governing variables (double-loop learning).
.claude/skills/yogsoth-ai-dialectical-reformulation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 24% | 0% |
Apply double-loop learning to question the problem's governing variables.
| Base SOP | Target | ±10% Range | |----------|--------|------------| | web-search | 20 | 18–22 | | web-research | 10 | 9–11 | | paper-overview | 30 | 27–33 | | paper-search | 20 | 18–22 | | paper-research | 10 | 9–11 |
<HARD-GATE>
| SOP | Done | Target | % |
|-----|------|--------|---|
| web-search | ? | 20 | ? |
| web-research | ? | 10 | ? |
| paper-overview | ? | 30 | ? |
| paper-search | ? | 20 | ? |
| paper-research | ? | 10 | ? |
Budget Gate: OPEN/CLOSED (>=80% required to exit)
</HARD-GATE>Import: web-search, web-research, paper-overview, paper-search, paper-research Subagent: governing-variable-surfacing, counter-assumption-generation
Surface Argyris governing variables (the unstated rules driving behavior), generate counter-assumptions for each, test whether the problem dissolves under alternative governing variables (double-loop learning).
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | deep-insight-dialectical-escalation | Double-loop learning escalation — surface governing variables, generate counter-assumptions, test if problem dissolves under alternatives, score wickedness if it persists. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | counter-assumption-generation | Generate dialectical opposites for governing variables — coherent alternative worldviews where the opposite is true. | | governing-variable-surfacing | Apply Argyris framework to identify governing variables — the unstated rules driving behavior in a research field. |
<!-- 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 | pass→pass | 17,344 | 38,698 | +123% | 1 | 1 | 0% | 2,661 | 6,720 | +153% | 0 | 0 | — |
case-02 | pass→pass | 15,369 | 20,636 | +34% | 1 | 1 | 0% | 2,483 | 3,874 | +56% | 0 | 0 | — |
case-03 | fail→pass | 17,516 | 2,950 | -83% | 1 | 1 | 0% | 1,243 | 1,052 | -15% | 0 | 0 | — |
case-04 | fail→pass | 11,220 | 2,795 | -75% | 1 | 1 | 0% | 1,995 | 1,061 | -47% | 0 | 0 | — |
case-05 | fail→pass | 11,230 | 3,394 | -70% | 1 | 1 | 0% | 1,966 | 1,273 | -35% | 0 | 0 | — |
case-06 | fail→pass | 12,528 | 3,161 | -75% | 1 | 1 | 0% | 873 | 1,179 | +35% | 0 | 0 | — |
case-07 | fail→pass | 11,197 | 8,882 | -21% | 1 | 1 | 0% | 1,677 | 2,083 | +24% | 0 | 0 | — |
case-08 | pass→fail | 10,462 | 4,867 | -53% | 1 | 1 | 0% | 1,696 | 1,444 | -15% | 0 | 0 | — |
case-09 | pass→pass | 15,646 | 6,155 | -61% | 1 | 1 | 0% | 2,471 | 1,634 | -34% | 0 | 0 | — |
case-10 | pass→fail | 20,238 | 33,065 | +63% | 1 | 1 | 0% | 3,895 | 6,719 | +73% | 0 | 0 | — |
case-11 | pass→fail | 16,876 | 23,256 | +38% | 1 | 1 | 0% | 3,087 | 4,445 | +44% | 0 | 0 | — |
case-16 | fail→pass | 6,033 | 3,090 | -49% | 1 | 1 | 0% | 891 | 1,044 | +17% | 0 | 0 | — |
case-12 | pass→pass | 7,895 | 17,906 | +127% | 1 | 1 | 0% | 1,522 | 4,154 | +173% | 0 | 0 | — |
case-13 | pass→pass | 10,577 | 27,155 | +157% | 1 | 1 | 0% | 1,593 | 3,969 | +149% | 0 | 0 | — |
case-14 | fail→pass | 9,587 | 1,949 | -80% | 1 | 1 | 0% | 1,540 | 891 | -42% | 0 | 0 | — |
case-15 | fail→pass | 11,547 | 2,185 | -81% | 1 | 1 | 0% | 1,734 | 883 | -49% | 0 | 0 | — |
case-17 | pass→pass | 13,760 | 17,323 | +26% | 1 | 1 | 0% | 2,208 | 3,189 | +44% | 0 | 0 | — |
case-18 | pass→fail | 9,035 | 10,502 | +16% | 1 | 1 | 0% | 1,474 | 1,305 | -11% | 0 | 0 | — |
case-19 | fail→fail | 8,435 | 7,523 | -11% | 1 | 1 | 0% | 1,262 | 1,863 | +48% | 0 | 0 | — |
case-20 | fail→pass | 31,781 | 9,355 | -71% | 1 | 1 | 0% | 1,066 | 2,263 | +112% | 0 | 0 | — |
case-21 | fail→pass | 13,944 | 3,366 | -76% | 1 | 1 | 0% | 2,147 | 1,109 | -48% | 0 | 0 | — |
case-22 | pass→fail | 12,867 | 20,708 | +61% | 1 | 1 | 0% | 2,381 | 4,369 | +83% | 0 | 0 | — |
case-23 | pass→fail | 12,946 | 3,801 | -71% | 1 | 1 | 0% | 2,106 | 1,127 | -46% | 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. 23 cases were attempted, and 19 counted toward the lift figure. The other 4 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 +9 percentage points is the difference between those two pass rates over the 19 comparable cases. 6 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.