Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use evolution mechanisms (selection, mutation, radiation) as design operators for generating and refining solution populations.
.claude/skills/yogsoth-ai-evolution-strategy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -55% | 0% |
Use evolution mechanisms (selection, mutation, radiation) as design operators.
| Resource | Target | Current | % | |----------|--------|---------|---| | web-search | 20 | 0 | 0% | | web-research | 8 | 0 | 0% | | paper-overview | 20 | 0 | 0% | | paper-search | 12 | 0 | 0% | | paper-research | 5 | 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 | |--------|------| | life-principles-application | Apply evolutionary principles as design constraints | | analogy-extraction | Extract transferable evolution mechanisms |
| SOP | Role | |-----|------| | evolution-mechanism-transfer | Map evolution mechanisms to design operations | | abstraction-to-design | Abstract evolutionary principles to design operators | | emulation-generation | Generate solutions using evolutionary operators | | biomimicry-synthesis | Synthesize final output |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | life-principles-application | Apply life's principles as design constraints. Orchestrates ecosystem-pattern-extraction → evolution-mechanism-transfer → abstraction-to-design. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | abstraction-to-design | Abstract biological principle to design principle. Bridge from biology to engineering. | | biomimicry-synthesis | Synthesize all biomimicry outputs into a structured idea report. Integrate biological strategies, design principles, and technical solutions. | | emulation-generation | Generate technical solutions emulating biological strategies. Bridge from design principle to concrete implementation. | | evolution-mechanism-transfer | Map evolution mechanisms to design operations. Translate selection, mutation, drift, radiation into design operators. |
<!-- 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-08 | fail→pass | 13,113 | 8,287 | -37% | 1 | 1 | 0% | 2,019 | 1,946 | -4% | 0 | 0 | — |
case-01 | pass→pass | 12,711 | 28,756 | +126% | 1 | 1 | 0% | 1,886 | 5,062 | +168% | 0 | 0 | — |
case-02 | fail→pass | 12,457 | 3,523 | -72% | 1 | 1 | 0% | 1,955 | 1,199 | -39% | 0 | 0 | — |
case-03 | fail→pass | 10,621 | 12,910 | +22% | 1 | 1 | 0% | 1,592 | 2,743 | +72% | 0 | 0 | — |
case-17 | pass→pass | 7,810 | 2,274 | -71% | 1 | 1 | 0% | 1,160 | 974 | -16% | 0 | 0 | — |
case-04 | fail→pass | 12,264 | 8,330 | -32% | 1 | 1 | 0% | 1,919 | 2,017 | +5% | 0 | 0 | — |
case-05 | fail→pass | 16,596 | 3,007 | -82% | 1 | 1 | 0% | 2,397 | 1,067 | -55% | 0 | 0 | — |
case-06 | pass→pass | 13,682 | 14,585 | +7% | 1 | 1 | 0% | 2,206 | 2,279 | +3% | 0 | 0 | — |
case-07 | pass→pass | 16,581 | 4,162 | -75% | 1 | 1 | 0% | 2,500 | 1,314 | -47% | 0 | 0 | — |
case-09 | fail→pass | 6,911 | 3,938 | -43% | 1 | 1 | 0% | 1,091 | 1,310 | +20% | 0 | 0 | — |
case-10 | fail→pass | 26,092 | 5,418 | -79% | 1 | 1 | 0% | 1,202 | 1,410 | +17% | 0 | 0 | — |
case-11 | pass→pass | 2,194 | 2,036 | -7% | 1 | 1 | 0% | 369 | 923 | +150% | 0 | 0 | — |
case-12 | fail→pass | 10,385 | 3,133 | -70% | 1 | 1 | 0% | 1,614 | 1,064 | -34% | 0 | 0 | — |
case-18 | fail→pass | 7,320 | 2,092 | -71% | 1 | 1 | 0% | 1,085 | 934 | -14% | 0 | 0 | — |
case-13 | fail→pass | 20,409 | 41,388 | +103% | 1 | 1 | 0% | 3,276 | 7,126 | +118% | 0 | 0 | — |
case-14 | pass→pass | 5,524 | 17,012 | +208% | 1 | 1 | 0% | 836 | 3,189 | +281% | 0 | 0 | — |
case-15 | fail→pass | 10,306 | 21,531 | +109% | 1 | 1 | 0% | 1,479 | 2,987 | +102% | 0 | 0 | — |
case-16 | pass→pass | 11,528 | 20,397 | +77% | 1 | 1 | 0% | 1,664 | 3,602 | +116% | 0 | 0 | — |
case-19 | fail→pass | 11,129 | 13,487 | +21% | 1 | 1 | 0% | 1,650 | 2,696 | +63% | 0 | 0 | — |
case-20 | fail→pass | 29,737 | 24,616 | -17% | 1 | 1 | 0% | 1,366 | 4,863 | +256% | 0 | 0 | — |
case-21 | pass→pass | 20,655 | 31,638 | +53% | 1 | 1 | 0% | 3,870 | 6,341 | +64% | 0 | 0 | — |
case-22 | pass→pass | 12,791 | 20,170 | +58% | 1 | 1 | 0% | 2,397 | 4,291 | +79% | 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 +59 percentage points is the difference between those two pass rates over the 20 comparable cases.
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.