Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Enumerate implicit assumptions in a problem statement or existing solution. Produces categorized assumption inventory (physical, social, temporal, economic, technical).
.claude/skills/yogsoth-ai-creative-ideation-assumption-surfacing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✓→✓ | ＝ Same ✓ | 27% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 55% | 0% |
| case-22 | ✓→✓ | ＝ Same ✓ | 66% | 0% |
| case-03 | ✗→✗ | ＝ Same ✗ | 41% | 0% |
| case-04 | ✗→✗ | ＝ Same ✗ | 3% | 0% |
Enumerate implicit assumptions in a problem or solution.
Subagent — spawned via subagent-spawning/spawn-agent skill.
Assumption surfacing requires systematic examination of every element in a problem statement, questioning what is taken for granted. Benefits from dedicated analytical focus without the distraction of solution generation.
<!-- 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-03 | fail→fail | 16,631 | 23,276 | +40% | 1 | 1 | 0% | 2,497 | 3,514 | +41% | 0 | 0 | — |
case-04 | fail→fail | 20,179 | 19,378 | -4% | 1 | 1 | 0% | 2,860 | 2,942 | +3% | 0 | 0 | — |
case-01 | fail→fail | 17,466 | 20,605 | +18% | 1 | 1 | 0% | 2,476 | 3,137 | +27% | 0 | 0 | — |
case-02 | fail→fail | 16,037 | 26,874 | +68% | 1 | 1 | 0% | 2,497 | 4,158 | +67% | 0 | 0 | — |
case-05 | fail→fail | 13,472 | 15,022 | +12% | 1 | 1 | 0% | 2,032 | 2,388 | +18% | 0 | 0 | — |
case-06 | fail→fail | 17,050 | 22,142 | +30% | 1 | 1 | 0% | 2,445 | 3,404 | +39% | 0 | 0 | — |
case-07 | fail→fail | 14,539 | 19,974 | +37% | 1 | 1 | 0% | 2,024 | 3,008 | +49% | 0 | 0 | — |
case-08 | fail→fail | 15,826 | 16,339 | +3% | 1 | 1 | 0% | 2,328 | 2,559 | +10% | 0 | 0 | — |
case-09 | fail→fail | 14,429 | 14,469 | +0% | 1 | 1 | 0% | 2,134 | 2,163 | +1% | 0 | 0 | — |
case-10 | fail→fail | 13,398 | 14,231 | +6% | 1 | 1 | 0% | 1,986 | 2,389 | +20% | 0 | 0 | — |
case-11 | fail→fail | 13,111 | 25,089 | +91% | 1 | 1 | 0% | 1,829 | 2,868 | +57% | 0 | 0 | — |
case-12 | fail→fail | 13,259 | 24,824 | +87% | 1 | 1 | 0% | 2,048 | 2,667 | +30% | 0 | 0 | — |
case-13 | fail→fail | 18,219 | 14,690 | -19% | 1 | 1 | 0% | 2,504 | 1,324 | -47% | 0 | 0 | — |
case-14 | fail→fail | 17,784 | 19,214 | +8% | 1 | 1 | 0% | 2,746 | 2,849 | +4% | 0 | 0 | — |
case-15 | fail→fail | 21,320 | 24,385 | +14% | 1 | 1 | 0% | 3,051 | 3,716 | +22% | 0 | 0 | — |
case-16 | fail→fail | 15,893 | 17,318 | +9% | 1 | 1 | 0% | 2,284 | 2,685 | +18% | 0 | 0 | — |
case-17 | fail→fail | 17,233 | 23,208 | +35% | 1 | 1 | 0% | 2,548 | 3,546 | +39% | 0 | 0 | — |
case-18 | fail→fail | 13,096 | 12,514 | -4% | 1 | 1 | 0% | 2,088 | 1,924 | -8% | 0 | 0 | — |
case-19 | fail→fail | 12,926 | 16,010 | +24% | 1 | 1 | 0% | 1,830 | 2,536 | +39% | 0 | 0 | — |
case-20 | pass→pass | 16,495 | 19,504 | +18% | 1 | 1 | 0% | 2,333 | 2,961 | +27% | 0 | 0 | — |
case-21 | pass→pass | 10,328 | 16,020 | +55% | 1 | 1 | 0% | 1,867 | 2,895 | +55% | 0 | 0 | — |
case-22 | pass→pass | 10,277 | 16,097 | +57% | 1 | 1 | 0% | 1,643 | 2,720 | +66% | 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 21 counted toward the lift figure. The other 1 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 -100 percentage points is the difference between those two pass rates over the 21 comparable cases. 2 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.