Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Multi-worldview comparison — CATWOE from multiple perspectives, reframing matrix across professional lenses, identify overlooked framings. Reveals what single-perspective analysis misses.
.claude/skills/yogsoth-ai-multi-worldview-comparison/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 47% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 518% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 309% | 0% |
| case-02 | ✓→✗ | ▼ Worse | 128% | 0% |
See what single-perspective analysis misses.
catwoe-analysis → reframing-matrix → appreciative-discovery (optional)
Subagent: catwoe-analysis, reframing-matrix, appreciative-discovery Shared: multi-stakeholder-simulation
Apply CATWOE (Customers, Actors, Transformation, Weltanschauung, Owner, Environment) from at least 4 stakeholder perspectives, then use reframing matrix (engineer, social scientist, artist, economist or domain-appropriate equivalents) to see what each profession would focus on.
Key insight: the problem looks fundamentally different from each perspective. The "real" problem is the intersection of all perspectives, not any single one.
<HARD-GATE>
- CATWOE analyses completed: >= 3 perspectives
- Reframing matrix perspectives: >= 4
- Overlooked framings identified: >= 2
- Synthesis of cross-perspective insights: >= 1
</HARD-GATE><!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | appreciative-discovery | Search for positive deviants and extract transferable principles using Appreciative Inquiry. | | catwoe-analysis | Apply Checkland's CATWOE analysis from a specific stakeholder perspective to reveal how the problem looks from that viewpoint. | | deep-insight-multi-stakeholder-simulation | Simulate multiple stakeholder perspectives evaluating a research gap, method, or proposal. Identifies blind spots from single-perspective analysis. | | reframing-matrix | Reframe the problem from 4 professional perspectives to reveal what each discipline would focus on. |
<!-- 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 | pass→pass | 17,991 | 41,478 | +131% | 1 | 1 | 0% | 1,990 | 6,301 | +217% | 0 | 0 | — |
case-02 | pass→fail | 24,698 | 47,703 | +93% | 1 | 1 | 0% | 3,719 | 8,489 | +128% | 0 | 0 | — |
case-03 | pass→pass | 20,580 | 33,954 | +65% | 1 | 1 | 0% | 2,497 | 5,876 | +135% | 0 | 0 | — |
case-04 | pass→pass | 23,220 | 26,911 | +16% | 1 | 1 | 0% | 2,778 | 4,631 | +67% | 0 | 0 | — |
case-05 | pass→pass | 24,058 | 9,762 | -59% | 1 | 1 | 0% | 3,085 | 1,284 | -58% | 0 | 0 | — |
case-06 | fail→pass | 15,031 | 24,761 | +65% | 1 | 1 | 0% | 1,700 | 2,492 | +47% | 0 | 0 | — |
case-07 | fail→pass | 24,728 | 36,696 | +48% | 1 | 1 | 0% | 893 | 5,523 | +518% | 0 | 0 | — |
case-08 | pass→pass | 22,675 | 32,922 | +45% | 1 | 1 | 0% | 2,608 | 4,885 | +87% | 0 | 0 | — |
case-09 | fail→pass | 19,587 | 7,687 | -61% | 1 | 1 | 0% | 2,231 | 918 | -59% | 0 | 0 | — |
case-10 | pass→pass | 13,158 | 21,639 | +64% | 1 | 1 | 0% | 1,326 | 3,940 | +197% | 0 | 0 | — |
case-11 | pass→pass | 11,818 | 30,802 | +161% | 1 | 1 | 0% | 1,093 | 5,411 | +395% | 0 | 0 | — |
case-12 | pass→pass | 5,490 | 6,707 | +22% | 1 | 1 | 0% | 883 | 1,543 | +75% | 0 | 0 | — |
case-13 | pass→fail | 19,591 | 7,898 | -60% | 1 | 1 | 0% | 2,108 | 934 | -56% | 0 | 0 | — |
case-14 | pass→fail | 14,810 | 2,567 | -83% | 1 | 1 | 0% | 1,646 | 870 | -47% | 0 | 0 | — |
case-15 | pass→pass | 20,355 | 8,573 | -58% | 1 | 1 | 0% | 2,243 | 1,001 | -55% | 0 | 0 | — |
case-16 | fail→pass | 13,077 | 29,715 | +127% | 1 | 1 | 0% | 1,092 | 4,461 | +309% | 0 | 0 | — |
case-17 | pass→pass | 17,978 | 42,820 | +138% | 1 | 1 | 0% | 2,512 | 4,044 | +61% | 0 | 0 | — |
case-18 | pass→pass | 14,249 | 38,211 | +168% | 1 | 1 | 0% | 2,144 | 5,642 | +163% | 0 | 0 | — |
case-19 | pass→pass | 15,877 | 22,542 | +42% | 1 | 1 | 0% | 2,187 | 3,538 | +62% | 0 | 0 | — |
case-20 | pass→pass | 14,212 | 25,005 | +76% | 1 | 1 | 0% | 2,175 | 4,410 | +103% | 0 | 0 | — |
case-21 | pass→pass | 14,730 | 20,937 | +42% | 1 | 1 | 0% | 2,325 | 4,100 | +76% | 0 | 0 | — |
case-22 | pass→pass | 13,621 | 30,549 | +124% | 1 | 1 | 0% | 2,172 | 5,179 | +138% | 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 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.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.