Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Synthesize diverse ideas into coherent solution concepts. Combines fragments from multiple ideation passes into structured, actionable ideas with clear mechanism descriptions.
.claude/skills/yogsoth-ai-idea-synthesis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-22 | ✓→✓ | ＝ Same ✓ | 38% | 0% |
| case-23 | ✓→✓ | ＝ Same ✓ | 35% | 0% |
| case-24 | ✓→✓ | ＝ Same ✓ | -6% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | 53% | 0% |
| case-02 | ✗→✗ | ＝ Same ✗ | 118% | 0% |
Synthesize diverse ideas into coherent solution concepts.
Subagent — spawned via subagent-spawning/spawn-agent skill.
Synthesis requires holding all intermediate outputs in context simultaneously and identifying connections, redundancies, and complementary fragments across multiple ideation passes.
<!-- 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 | 12,257 | 14,662 | +20% | 1 | 1 | 0% | 1,048 | 1,606 | +53% | 0 | 0 | — |
case-02 | fail→fail | 12,442 | 19,341 | +55% | 1 | 1 | 0% | 1,102 | 2,397 | +118% | 0 | 0 | — |
case-03 | fail→fail | 11,286 | 23,138 | +105% | 1 | 1 | 0% | 957 | 2,839 | +197% | 0 | 0 | — |
case-04 | fail→fail | 26,059 | 52,778 | +103% | 1 | 1 | 0% | 3,003 | 7,738 | +158% | 0 | 0 | — |
case-05 | fail→fail | 14,171 | 41,222 | +191% | 1 | 1 | 0% | 1,299 | 6,244 | +381% | 0 | 0 | — |
case-06 | fail→fail | 21,684 | 33,025 | +52% | 1 | 1 | 0% | 2,459 | 4,276 | +74% | 0 | 0 | — |
case-07 | fail→fail | 23,367 | 37,710 | +61% | 1 | 1 | 0% | 2,707 | 3,238 | +20% | 0 | 0 | — |
case-08 | fail→fail | 24,907 | 24,782 | -1% | 1 | 1 | 0% | 2,873 | 967 | -66% | 0 | 0 | — |
case-09 | fail→fail | 25,475 | 37,632 | +48% | 1 | 1 | 0% | 3,194 | 5,384 | +69% | 0 | 0 | — |
case-10 | fail→fail | 24,350 | 37,784 | +55% | 1 | 1 | 0% | 2,792 | 5,068 | +82% | 0 | 0 | — |
case-11 | fail→fail | 27,435 | 75,786 | +176% | 1 | 1 | 0% | 3,435 | 4,963 | +44% | 0 | 0 | — |
case-12 | fail→fail | 21,667 | 40,646 | +88% | 1 | 1 | 0% | 2,498 | 5,978 | +139% | 0 | 0 | — |
case-13 | fail→fail | 13,051 | 51,220 | +292% | 1 | 1 | 0% | 1,151 | 7,739 | +572% | 0 | 0 | — |
case-14 | fail→fail | 20,961 | 34,314 | +64% | 1 | 1 | 0% | 2,402 | 813 | -66% | 0 | 0 | — |
case-15 | fail→fail | 23,759 | 44,438 | +87% | 1 | 1 | 0% | 2,814 | 5,994 | +113% | 0 | 0 | — |
case-16 | fail→fail | 25,819 | 31,017 | +20% | 1 | 1 | 0% | 2,981 | 3,923 | +32% | 0 | 0 | — |
case-17 | fail→fail | 20,764 | 32,925 | +59% | 1 | 1 | 0% | 2,409 | 4,620 | +92% | 0 | 0 | — |
case-18 | fail→fail | 24,724 | 41,940 | +70% | 1 | 1 | 0% | 3,035 | 6,210 | +105% | 0 | 0 | — |
case-19 | fail→fail | 24,088 | 23,267 | -3% | 1 | 1 | 0% | 2,801 | 1,061 | -62% | 0 | 0 | — |
case-20 | fail→fail | 22,199 | 41,145 | +85% | 1 | 1 | 0% | 2,572 | 6,112 | +138% | 0 | 0 | — |
case-21 | fail→fail | 25,421 | 43,334 | +70% | 1 | 1 | 0% | 2,886 | 4,505 | +56% | 0 | 0 | — |
case-22 | pass→pass | 20,400 | 24,487 | +20% | 1 | 1 | 0% | 2,607 | 3,601 | +38% | 0 | 0 | — |
case-23 | pass→pass | 44,501 | 54,764 | +23% | 1 | 1 | 0% | 6,174 | 8,356 | +35% | 0 | 0 | — |
case-24 | pass→pass | 26,031 | 24,489 | -6% | 1 | 1 | 0% | 3,001 | 2,835 | -6% | 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. 24 cases were attempted, and 21 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 21 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.