Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Find domain's unquestioned beliefs. Systematic identification of dogma that constrains innovation.
.claude/skills/yogsoth-ai-sacred-cow-identification/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✓→✓ | ＝ Same ✓ | 54% | 0% |
| case-19 | ✓→✓ | ＝ Same ✓ | 23% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 70% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 9% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | 29% | 0% |
Find domain's unquestioned beliefs.
Subagent — spawned via subagent-spawning/spawn-agent skill.
Sacred cow identification requires deep domain analysis and the ability to distinguish between well-justified constraints and mere tradition. Benefits from dedicated critical thinking without solution-generation pressure.
<!-- 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 | 29,929 | 35,799 | +20% | 1 | 1 | 0% | 3,653 | 4,722 | +29% | 0 | 0 | — |
case-02 | fail→fail | 14,630 | 65,702 | +349% | 1 | 1 | 0% | 1,360 | 5,500 | +304% | 0 | 0 | — |
case-03 | fail→fail | 26,543 | 54,718 | +106% | 1 | 1 | 0% | 3,218 | 7,124 | +121% | 0 | 0 | — |
case-04 | fail→fail | 28,755 | 35,901 | +25% | 1 | 1 | 0% | 3,395 | 4,728 | +39% | 0 | 0 | — |
case-05 | fail→fail | 20,345 | 19,507 | -4% | 1 | 1 | 0% | 2,239 | 3,089 | +38% | 0 | 0 | — |
case-06 | fail→fail | 22,141 | 30,320 | +37% | 1 | 1 | 0% | 2,571 | 4,070 | +58% | 0 | 0 | — |
case-07 | fail→fail | 21,458 | 27,039 | +26% | 1 | 1 | 0% | 2,504 | 3,453 | +38% | 0 | 0 | — |
case-08 | fail→fail | 20,012 | 44,349 | +122% | 1 | 1 | 0% | 2,829 | 5,834 | +106% | 0 | 0 | — |
case-09 | fail→fail | 21,773 | 42,481 | +95% | 1 | 1 | 0% | 3,224 | 6,361 | +97% | 0 | 0 | — |
case-10 | fail→fail | 20,764 | 43,820 | +111% | 1 | 1 | 0% | 2,980 | 4,111 | +38% | 0 | 0 | — |
case-11 | fail→fail | 22,126 | 30,462 | +38% | 1 | 1 | 0% | 2,496 | 4,110 | +65% | 0 | 0 | — |
case-12 | fail→fail | 23,973 | 38,058 | +59% | 1 | 1 | 0% | 3,049 | 4,847 | +59% | 0 | 0 | — |
case-13 | fail→fail | 29,020 | 29,237 | +1% | 1 | 1 | 0% | 3,558 | 4,454 | +25% | 0 | 0 | — |
case-14 | fail→fail | 23,071 | 35,551 | +54% | 1 | 1 | 0% | 2,581 | 4,564 | +77% | 0 | 0 | — |
case-15 | fail→fail | 28,679 | 29,001 | +1% | 1 | 1 | 0% | 3,320 | 3,506 | +6% | 0 | 0 | — |
case-16 | fail→fail | 21,194 | 37,167 | +75% | 1 | 1 | 0% | 2,989 | 5,472 | +83% | 0 | 0 | — |
case-17 | fail→fail | 24,744 | 41,059 | +66% | 1 | 1 | 0% | 2,824 | 5,619 | +99% | 0 | 0 | — |
case-18 | pass→pass | 21,412 | 28,093 | +31% | 1 | 1 | 0% | 2,487 | 3,825 | +54% | 0 | 0 | — |
case-19 | pass→pass | 18,925 | 21,535 | +14% | 1 | 1 | 0% | 2,679 | 3,307 | +23% | 0 | 0 | — |
case-20 | pass→pass | 13,269 | 18,624 | +40% | 1 | 1 | 0% | 1,542 | 2,620 | +70% | 0 | 0 | — |
case-21 | pass→pass | 18,375 | 13,557 | -26% | 1 | 1 | 0% | 2,093 | 2,276 | +9% | 0 | 0 | — |
case-22 | fail→fail | 23,132 | 34,971 | +51% | 1 | 1 | 0% | 2,668 | 4,606 | +73% | 0 | 0 | — |
case-23 | fail→fail | 19,329 | 26,714 | +38% | 1 | 1 | 0% | 2,087 | 3,332 | +60% | 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. 23 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Other measured skills in the registry, with their headline benchmark lift.