Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Run blend as mental simulation
.claude/skills/yogsoth-ai-blend-elaboration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✓→✓ | ＝ Same ✓ | 20% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 82% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 111% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | -82% | 0% |
| case-02 | ✗→✗ | ＝ Same ✗ | 124% | 0% |
Run blend as mental simulation — let the blend "run" according to its own logic to discover consequences, implications, and additional emergent properties.
Subagent — spawned via subagent-spawning/spawn-agent skill.
Blend elaboration requires imaginative simulation — running the blend forward in time, exploring consequences, and discovering properties that only emerge through dynamic interaction. Benefits from sustained imaginative focus.
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | spawn-agent | Spawn a customized CC subagent with full MCP tool access. Used by SOPs that declare execution: subagent. |
<!-- 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 | 30,722 | 28,339 | -8% | 1 | 1 | 0% | 3,577 | 657 | -82% | 0 | 0 | — |
case-02 | fail→fail | 25,032 | 54,715 | +119% | 1 | 1 | 0% | 3,333 | 7,475 | +124% | 0 | 0 | — |
case-03 | fail→fail | 35,053 | 28,767 | -18% | 1 | 1 | 0% | 4,456 | 1,414 | -68% | 0 | 0 | — |
case-04 | pass→pass | 14,109 | 24,319 | +72% | 1 | 1 | 0% | 2,355 | 2,817 | +20% | 0 | 0 | — |
case-05 | pass→pass | 18,971 | 27,329 | +44% | 1 | 1 | 0% | 2,295 | 4,186 | +82% | 0 | 0 | — |
case-06 | pass→pass | 9,478 | 20,853 | +120% | 1 | 1 | 0% | 1,755 | 3,711 | +111% | 0 | 0 | — |
case-07 | fail→fail | 22,457 | 22,371 | -0% | 1 | 1 | 0% | 2,400 | 3,996 | +67% | 0 | 0 | — |
case-08 | fail→fail | 13,667 | 18,731 | +37% | 1 | 1 | 0% | 1,904 | 2,831 | +49% | 0 | 0 | — |
case-09 | fail→fail | 22,628 | 45,707 | +102% | 1 | 1 | 0% | 3,113 | 4,947 | +59% | 0 | 0 | — |
case-10 | fail→fail | 19,289 | 49,670 | +158% | 1 | 1 | 0% | 2,858 | 4,191 | +47% | 0 | 0 | — |
case-11 | fail→fail | 15,347 | 14,653 | -5% | 1 | 1 | 0% | 1,315 | 1,357 | +3% | 0 | 0 | — |
case-12 | fail→fail | 32,933 | 37,039 | +12% | 1 | 1 | 0% | 3,799 | 4,972 | +31% | 0 | 0 | — |
case-13 | fail→fail | 26,997 | 32,640 | +21% | 1 | 1 | 0% | 3,200 | 4,552 | +42% | 0 | 0 | — |
case-14 | fail→fail | 27,873 | 40,757 | +46% | 1 | 1 | 0% | 3,203 | 5,295 | +65% | 0 | 0 | — |
case-15 | fail→fail | 11,854 | 20,384 | +72% | 1 | 1 | 0% | 1,833 | 3,348 | +83% | 0 | 0 | — |
case-16 | fail→fail | 22,634 | 44,142 | +95% | 1 | 1 | 0% | 3,612 | 6,508 | +80% | 0 | 0 | — |
case-17 | fail→fail | 23,550 | 43,280 | +84% | 1 | 1 | 0% | 2,998 | 4,211 | +40% | 0 | 0 | — |
case-18 | fail→fail | 24,313 | 32,144 | +32% | 1 | 1 | 0% | 2,939 | 1,204 | -59% | 0 | 0 | — |
case-19 | fail→fail | 29,825 | 46,801 | +57% | 1 | 1 | 0% | 3,442 | 4,265 | +24% | 0 | 0 | — |
case-20 | fail→fail | 29,838 | 39,560 | +33% | 1 | 1 | 0% | 4,264 | 5,382 | +26% | 0 | 0 | — |
case-21 | fail→fail | 21,826 | 49,935 | +129% | 1 | 1 | 0% | 3,143 | 5,109 | +63% | 0 | 0 | — |
case-22 | fail→fail | 25,446 | 33,661 | +32% | 1 | 1 | 0% | 3,180 | 4,740 | +49% | 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 19 counted toward the lift figure. The other 3 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 0 percentage points is the difference between those two pass rates over the 19 comparable cases.
Other measured skills in the registry, with their headline benchmark lift.