Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →CCA: pairwise consistency checking to reduce solution space 90-99%
.claude/skills/yogsoth-ai-cross-consistency-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 97% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 202% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 46% | 0% |
Pairwise consistency checking (CCA) to reduce the total morphological solution space by 90-99%, leaving only internally consistent configurations.
| Resource | Target | Current | % | |----------|--------|---------|---| | web-search | 15 | 0 | 0% | | web-research | 5 | 0 | 0% | | paper-overview | 15 | 0 | 0% | | paper-search | 10 | 0 | 0% | | paper-research | 3 | 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 | |--------|------| | consistency-checking | Core tactic — evaluate and reduce |
| SOP | Role | |-----|------| | consistency-pair-evaluation | Evaluate pairwise consistency judgments | | solution-space-reduction | Apply CCA to remove inconsistent combinations | | morphological-synthesis | Synthesize reduced space report |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | creative-ideation-consistency-checking | Pairwise consistency evaluation to reduce solution space by identifying and removing inconsistent combinations. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | creative-ideation-consistency-pair-evaluation | Evaluate pairwise value consistency (logical/empirical/normative) | | morphological-synthesis | Synthesize all morphological exploration outputs | | solution-space-reduction | Apply CCA to remove inconsistent combinations |
<!-- 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→fail | 31,961 | 7,818 | -76% | 1 | 1 | 0% | 6,233 | 899 | -86% | 0 | 0 | — |
case-02 | fail→fail | 28,286 | 25,844 | -9% | 1 | 1 | 0% | 5,810 | 6,736 | +16% | 0 | 0 | — |
case-03 | pass→fail | 15,557 | 37,146 | +139% | 1 | 1 | 0% | 2,635 | 6,714 | +155% | 0 | 0 | — |
case-12 | fail→fail | 19,768 | 9,829 | -50% | 1 | 1 | 0% | 3,564 | 1,277 | -64% | 0 | 0 | — |
case-04 | pass→fail | 12,692 | 11,559 | -9% | 1 | 1 | 0% | 2,362 | 1,128 | -52% | 0 | 0 | — |
case-05 | pass→fail | 18,542 | 10,846 | -42% | 1 | 1 | 0% | 3,369 | 1,333 | -60% | 0 | 0 | — |
case-06 | fail→pass | 19,691 | 32,788 | +67% | 1 | 1 | 0% | 3,406 | 6,702 | +97% | 0 | 0 | — |
case-07 | fail→fail | 43,940 | 62,115 | +41% | 1 | 1 | 0% | 1,812 | 1,947 | +7% | 0 | 0 | — |
case-21 | fail→fail | 21,378 | 33,524 | +57% | 1 | 1 | 0% | 3,887 | 6,448 | +66% | 0 | 0 | — |
case-08 | fail→pass | 15,613 | 9,733 | -38% | 1 | 1 | 0% | 2,391 | 2,031 | -15% | 0 | 0 | — |
case-09 | fail→pass | 55,765 | 33,648 | -40% | 1 | 1 | 0% | 2,215 | 6,693 | +202% | 0 | 0 | — |
case-10 | pass→pass | 10,189 | 4,550 | -55% | 1 | 1 | 0% | 1,704 | 1,252 | -27% | 0 | 0 | — |
case-11 | pass→pass | 31,857 | 42,214 | +33% | 1 | 1 | 0% | 2,712 | 4,266 | +57% | 0 | 0 | — |
case-13 | fail→fail | 60,550 | 32,566 | -46% | 1 | 1 | 0% | 6,180 | 6,699 | +8% | 0 | 0 | — |
case-14 | fail→pass | 13,525 | 9,053 | -33% | 1 | 1 | 0% | 2,345 | 1,935 | -17% | 0 | 0 | — |
case-15 | fail→pass | 27,043 | 31,777 | +18% | 1 | 1 | 0% | 4,597 | 6,694 | +46% | 0 | 0 | — |
case-16 | fail→pass | 25,334 | 113,524 | +348% | 1 | 1 | 0% | 4,129 | 6,688 | +62% | 0 | 0 | — |
case-17 | fail→fail | 17,787 | 33,561 | +89% | 1 | 1 | 0% | 3,050 | 6,700 | +120% | 0 | 0 | — |
case-18 | pass→pass | 16,929 | 61,314 | +262% | 1 | 1 | 0% | 2,713 | 5,487 | +102% | 0 | 0 | — |
case-19 | fail→fail | 32,334 | 7,866 | -76% | 1 | 1 | 0% | 2,258 | 1,129 | -50% | 0 | 0 | — |
case-20 | pass→pass | 9,994 | 8,696 | -13% | 1 | 1 | 0% | 1,549 | 1,801 | +16% | 0 | 0 | — |
case-22 | fail→pass | 19,723 | 38,969 | +98% | 1 | 1 | 0% | 3,188 | 6,680 | +110% | 0 | 0 | — |
case-23 | fail→fail | 36,017 | 36,870 | +2% | 1 | 1 | 0% | 1,318 | 1,078 | -18% | 0 | 0 | — |
case-24 | fail→pass | 26,870 | 45,566 | +70% | 1 | 1 | 0% | 4,439 | 6,681 | +51% | 0 | 0 | — |
case-25 | pass→fail | 38,444 | 58,778 | +53% | 1 | 1 | 0% | 3,458 | 1,028 | -70% | 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, and 15 counted toward the lift figure. The other 10 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 +16 percentage points is the difference between those two pass rates over the 15 comparable cases. 4 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.