Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute SCAMPER 7 operators on a target solution. Subagent self-selects best 2-3 operators for deepest exploration.
.claude/skills/yogsoth-ai-scamper-divergence/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✓→✗ | ▼ Worse | 209% | 0% |
| case-22 | ✓→✗ | ▼ Worse | 31% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 97% | 0% |
| case-13 | ✗→✗ | ＝ Same ✗ | 64% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | -76% | 0% |
Execute all 7 SCAMPER operators and self-select the most productive 2-3 for deep exploration.
Subagent — spawned via subagent-spawning/spawn-agent skill.
SCAMPER requires systematic application of 7 distinct mental operators, each demanding a different creative lens. Benefits from dedicated context that can fully commit to each operator without cross-contamination.
<!-- 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-13 | fail→fail | 38,227 | 42,302 | +11% | 1 | 1 | 0% | 3,357 | 5,515 | +64% | 0 | 0 | — |
case-01 | fail→fail | 36,834 | 24,809 | -33% | 1 | 1 | 0% | 4,604 | 1,104 | -76% | 0 | 0 | — |
case-12 | fail→fail | 35,680 | 40,065 | +12% | 1 | 1 | 0% | 4,585 | 5,457 | +19% | 0 | 0 | — |
case-02 | fail→fail | 34,219 | 20,359 | -41% | 1 | 1 | 0% | 5,017 | 1,403 | -72% | 0 | 0 | — |
case-03 | fail→fail | 33,824 | 48,903 | +45% | 1 | 1 | 0% | 4,258 | 7,282 | +71% | 0 | 0 | — |
case-04 | fail→fail | 30,807 | 23,062 | -25% | 1 | 1 | 0% | 3,749 | 899 | -76% | 0 | 0 | — |
case-05 | fail→fail | 21,044 | 69,222 | +229% | 1 | 1 | 0% | 3,189 | 7,303 | +129% | 0 | 0 | — |
case-06 | fail→fail | 31,741 | 55,935 | +76% | 1 | 1 | 0% | 3,883 | 8,114 | +109% | 0 | 0 | — |
case-07 | fail→fail | 33,221 | 26,608 | -20% | 1 | 1 | 0% | 4,108 | 1,242 | -70% | 0 | 0 | — |
case-08 | fail→fail | 33,800 | 54,871 | +62% | 1 | 1 | 0% | 4,273 | 8,401 | +97% | 0 | 0 | — |
case-09 | fail→fail | 25,225 | 56,905 | +126% | 1 | 1 | 0% | 3,285 | 7,543 | +130% | 0 | 0 | — |
case-10 | fail→fail | 30,205 | 62,377 | +107% | 1 | 1 | 0% | 3,690 | 8,694 | +136% | 0 | 0 | — |
case-11 | fail→fail | 31,262 | 31,818 | +2% | 1 | 1 | 0% | 4,421 | 1,330 | -70% | 0 | 0 | — |
case-14 | fail→fail | 44,974 | 70,579 | +57% | 1 | 1 | 0% | 4,362 | 8,399 | +93% | 0 | 0 | — |
case-15 | fail→fail | 42,690 | 28,785 | -33% | 1 | 1 | 0% | 4,585 | 1,581 | -66% | 0 | 0 | — |
case-16 | fail→fail | 31,155 | 50,119 | +61% | 1 | 1 | 0% | 3,913 | 7,109 | +82% | 0 | 0 | — |
case-17 | fail→fail | 59,547 | 69,806 | +17% | 1 | 1 | 0% | 8,219 | 8,396 | +2% | 0 | 0 | — |
case-18 | fail→fail | 25,164 | 19,993 | -21% | 1 | 1 | 0% | 3,900 | 739 | -81% | 0 | 0 | — |
case-19 | fail→fail | 27,756 | 57,693 | +108% | 1 | 1 | 0% | 4,115 | 6,611 | +61% | 0 | 0 | — |
case-20 | pass→pass | 23,110 | 40,323 | +74% | 1 | 1 | 0% | 3,448 | 6,781 | +97% | 0 | 0 | — |
case-21 | pass→fail | 17,540 | 54,088 | +208% | 1 | 1 | 0% | 2,669 | 8,236 | +209% | 0 | 0 | — |
case-22 | pass→fail | 28,498 | 38,383 | +35% | 1 | 1 | 0% | 4,992 | 6,556 | +31% | 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 15 counted toward the lift figure. The other 7 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 -9 percentage points is the difference between those two pass rates over the 15 comparable cases. 9 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.