Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Reassemble decomposed fragments into novel structures through systematic recombination of components.
.claude/skills/yogsoth-ai-recombination-architecture/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 210% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 78% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 192% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 18% | 0% |
Take decomposed system fragments and reassemble them into novel structural arrangements.
| Resource | Target | Current | % | |----------|--------|---------|---| | web-search | 20 | 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 | |--------|------| | combination-mapping | Enumerate recombination possibilities | | component-decomposition | Provide component inventory for recombination |
| SOP | Role | |-----|------| | recombination-generation | Generate novel combinations from fragments | | function-model-construction | Model functions to guide viable recombinations | | structural-synthesis | Synthesize recombination outcomes |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | component-decomposition | Decompose system into functional components, identify dependencies, and surface trimming candidates. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | recombination-generation | Reassemble decomposed system fragments into novel structural arrangements that create emergent value. | | structural-synthesis | Synthesize all structural transformation outputs into a coherent, ranked idea report with lineage tracking. |
<!-- 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 | 33,011 | 17,928 | -46% | 1 | 1 | 0% | 4,660 | 1,666 | -64% | 0 | 0 | — |
case-07 | fail→pass | 19,618 | 55,610 | +183% | 1 | 1 | 0% | 2,816 | 8,737 | +210% | 0 | 0 | — |
case-02 | fail→fail | 50,564 | 54,108 | +7% | 1 | 1 | 0% | 8,264 | 8,767 | +6% | 0 | 0 | — |
case-03 | fail→fail | 44,478 | 38,292 | -14% | 1 | 1 | 0% | 7,094 | 7,017 | -1% | 0 | 0 | — |
case-04 | pass→pass | 31,494 | 51,163 | +62% | 1 | 1 | 0% | 4,123 | 8,742 | +112% | 0 | 0 | — |
case-05 | pass→pass | 36,953 | 41,692 | +13% | 1 | 1 | 0% | 5,699 | 8,731 | +53% | 0 | 0 | — |
case-06 | pass→fail | 22,939 | 44,309 | +93% | 1 | 1 | 0% | 3,599 | 8,733 | +143% | 0 | 0 | — |
case-08 | fail→pass | 18,813 | 24,186 | +29% | 1 | 1 | 0% | 2,074 | 3,699 | +78% | 0 | 0 | — |
case-09 | pass→pass | 33,902 | 54,176 | +60% | 1 | 1 | 0% | 3,658 | 8,736 | +139% | 0 | 0 | — |
case-10 | fail→fail | 24,263 | 56,679 | +134% | 1 | 1 | 0% | 3,931 | 8,168 | +108% | 0 | 0 | — |
case-11 | fail→pass | 19,019 | 53,029 | +179% | 1 | 1 | 0% | 2,266 | 6,607 | +192% | 0 | 0 | — |
case-12 | pass→pass | 39,916 | 61,974 | +55% | 1 | 1 | 0% | 5,940 | 7,850 | +32% | 0 | 0 | — |
case-13 | pass→pass | 25,201 | 49,566 | +97% | 1 | 1 | 0% | 3,208 | 7,820 | +144% | 0 | 0 | — |
case-14 | fail→fail | 38,160 | 54,055 | +42% | 1 | 1 | 0% | 6,236 | 8,739 | +40% | 0 | 0 | — |
case-15 | pass→pass | 34,782 | 63,914 | +84% | 1 | 1 | 0% | 5,388 | 8,654 | +61% | 0 | 0 | — |
case-16 | pass→fail | 18,239 | 67,499 | +270% | 1 | 1 | 0% | 2,837 | 8,723 | +207% | 0 | 0 | — |
case-17 | fail→pass | 40,680 | 7,096 | -83% | 1 | 1 | 0% | 827 | 782 | -5% | 0 | 0 | — |
case-18 | fail→pass | 12,423 | 7,266 | -42% | 1 | 1 | 0% | 736 | 868 | +18% | 0 | 0 | — |
case-19 | fail→pass | 7,929 | 6,692 | -16% | 1 | 1 | 0% | 433 | 743 | +72% | 0 | 0 | — |
case-20 | fail→pass | 23,668 | 6,763 | -71% | 1 | 1 | 0% | 392 | 731 | +86% | 0 | 0 | — |
case-21 | fail→pass | 10,338 | 6,888 | -33% | 1 | 1 | 0% | 725 | 776 | +7% | 0 | 0 | — |
case-22 | fail→pass | 18,560 | 28,884 | +56% | 1 | 1 | 0% | 2,121 | 4,155 | +96% | 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, and 20 counted toward the lift figure. The other 2 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 +32 percentage points is the difference between those two pass rates over the 20 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.