Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Simulates diverse stakeholder perspectives and their strongest objections/support arguments. Shared across steel-manning and consensus campaigns.
.claude/skills/yogsoth-ai-convergence-multi-stakeholder-simulation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✓→✓ | ＝ Same ✓ | 40% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 43% | 0% |
| case-22 | ✓→✓ | ＝ Same ✓ | 23% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | 84% | 0% |
| case-02 | ✗→✗ | ＝ Same ✗ | 56% | 0% |
Simulates multiple stakeholder perspectives to surface objections and support arguments that a single-perspective analysis would miss.
Subagent — spawned via subagent-spawning/spawn-agent skill.
Stakeholder simulation requires genuine perspective-taking — inhabiting each role fully without contamination from other perspectives. Dedicated context enables authentic role-play.
Must simulate ≥3 distinct perspectives. Fewer than 3 perspectives cannot claim multi-stakeholder coverage.
<!-- 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 | 20,231 | 33,712 | +67% | 1 | 1 | 0% | 2,943 | 5,412 | +84% | 0 | 0 | — |
case-02 | fail→fail | 20,836 | 30,149 | +45% | 1 | 1 | 0% | 3,044 | 4,749 | +56% | 0 | 0 | — |
case-03 | fail→fail | 17,124 | 23,101 | +35% | 1 | 1 | 0% | 2,738 | 3,538 | +29% | 0 | 0 | — |
case-04 | fail→fail | 10,789 | 18,702 | +73% | 1 | 1 | 0% | 1,603 | 2,852 | +78% | 0 | 0 | — |
case-05 | fail→fail | 6,596 | 10,514 | +59% | 1 | 1 | 0% | 986 | 1,726 | +75% | 0 | 0 | — |
case-06 | fail→fail | 18,130 | 44,168 | +144% | 1 | 1 | 0% | 2,724 | 5,213 | +91% | 0 | 0 | — |
case-07 | fail→fail | 17,461 | 27,871 | +60% | 1 | 1 | 0% | 2,576 | 4,114 | +60% | 0 | 0 | — |
case-08 | fail→fail | 20,591 | 22,289 | +8% | 1 | 1 | 0% | 2,946 | 3,523 | +20% | 0 | 0 | — |
case-09 | fail→fail | 18,966 | 31,554 | +66% | 1 | 1 | 0% | 2,647 | 3,999 | +51% | 0 | 0 | — |
case-10 | fail→fail | 18,916 | 27,929 | +48% | 1 | 1 | 0% | 2,674 | 4,387 | +64% | 0 | 0 | — |
case-11 | fail→fail | 11,746 | 18,606 | +58% | 1 | 1 | 0% | 1,698 | 3,089 | +82% | 0 | 0 | — |
case-12 | fail→fail | 19,092 | 33,479 | +75% | 1 | 1 | 0% | 2,796 | 4,828 | +73% | 0 | 0 | — |
case-13 | fail→fail | 17,162 | 32,557 | +90% | 1 | 1 | 0% | 2,490 | 4,773 | +92% | 0 | 0 | — |
case-14 | fail→fail | 18,597 | 26,323 | +42% | 1 | 1 | 0% | 2,654 | 3,904 | +47% | 0 | 0 | — |
case-15 | fail→fail | 17,883 | 27,091 | +51% | 1 | 1 | 0% | 2,565 | 4,052 | +58% | 0 | 0 | — |
case-16 | fail→fail | 17,441 | 23,598 | +35% | 1 | 1 | 0% | 2,508 | 3,537 | +41% | 0 | 0 | — |
case-17 | fail→fail | 17,449 | 29,876 | +71% | 1 | 1 | 0% | 2,485 | 4,576 | +84% | 0 | 0 | — |
case-18 | fail→fail | 13,597 | 31,894 | +135% | 1 | 1 | 0% | 1,972 | 4,970 | +152% | 0 | 0 | — |
case-19 | fail→fail | 17,336 | 102,783 | +493% | 1 | 1 | 0% | 2,397 | 4,789 | +100% | 0 | 0 | — |
case-20 | pass→pass | 19,871 | 27,784 | +40% | 1 | 1 | 0% | 3,093 | 4,326 | +40% | 0 | 0 | — |
case-21 | pass→pass | 11,988 | 19,485 | +63% | 1 | 1 | 0% | 2,381 | 3,411 | +43% | 0 | 0 | — |
case-22 | pass→pass | 12,076 | 15,731 | +30% | 1 | 1 | 0% | 2,701 | 3,310 | +23% | 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. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases.
Other measured skills in the registry, with their headline benchmark lift.