Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Systematically expose hidden system boundaries — CSH 12-question is/ought comparison, identify excluded stakeholders, reveal blind spots. Combines csh-12-question, jtbd-mapping, and salience-classification SOPs.
.claude/skills/yogsoth-ai-boundary-unfolding/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 19% | 0% |
Expose hidden system boundaries and excluded perspectives.
Subagent: csh-12-question, jtbd-mapping, salience-classification Shared: multi-stakeholder-simulation
Apply CSH 12 questions (4 dimensions × 3 questions, is vs ought). Map stakeholder jobs. Classify by salience model. Identify who is systematically excluded from the current framing.
<HARD-GATE>
- CSH matrix: >= 1 completed
- stakeholders mapped: >= 4
- salience classifications: >= 4
- excluded perspectives identified: >= 1
</HARD-GATE><!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | csh-12-question | Apply Ulrich's Critical Systems Heuristics 12 questions across 4 dimensions (motivation, control, expertise, legitimacy) comparing is vs ought. | | deep-insight-multi-stakeholder-simulation | Simulate multiple stakeholder perspectives evaluating a research gap, method, or proposal. Identifies blind spots from single-perspective analysis. | | jtbd-mapping | Map stakeholder Jobs-to-be-Done — functional, emotional, and social jobs for each affected party. Identifies unserved jobs as opportunity signals. | | salience-classification | Classify stakeholders by Mitchell et al. framework (Power, Legitimacy, Urgency). Assigns salience category and identifies systematically excluded parties. |
<!-- 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-02 | fail→pass | 25,137 | 26,327 | +5% | 1 | 1 | 0% | 3,849 | 4,385 | +14% | 0 | 0 | — |
case-01 | pass→pass | 26,601 | 22,703 | -15% | 1 | 1 | 0% | 4,075 | 3,885 | -5% | 0 | 0 | — |
case-03 | pass→fail | 15,804 | 31,787 | +101% | 1 | 1 | 0% | 2,628 | 5,480 | +109% | 0 | 0 | — |
case-04 | pass→fail | 18,565 | 29,111 | +57% | 1 | 1 | 0% | 2,596 | 4,632 | +78% | 0 | 0 | — |
case-05 | pass→pass | 14,121 | 27,030 | +91% | 1 | 1 | 0% | 2,589 | 5,082 | +96% | 0 | 0 | — |
case-06 | pass→pass | 12,320 | 23,529 | +91% | 1 | 1 | 0% | 1,940 | 3,936 | +103% | 0 | 0 | — |
case-07 | fail→pass | 22,052 | 29,205 | +32% | 1 | 1 | 0% | 3,042 | 4,612 | +52% | 0 | 0 | — |
case-08 | pass→pass | 22,122 | 26,137 | +18% | 1 | 1 | 0% | 3,500 | 4,331 | +24% | 0 | 0 | — |
case-09 | pass→pass | 16,018 | 30,288 | +89% | 1 | 1 | 0% | 2,627 | 4,973 | +89% | 0 | 0 | — |
case-10 | fail→pass | 21,491 | 28,773 | +34% | 1 | 1 | 0% | 3,274 | 4,621 | +41% | 0 | 0 | — |
case-11 | fail→pass | 29,849 | 25,446 | -15% | 1 | 1 | 0% | 4,472 | 4,210 | -6% | 0 | 0 | — |
case-12 | pass→pass | 19,543 | 29,896 | +53% | 1 | 1 | 0% | 2,650 | 4,910 | +85% | 0 | 0 | — |
case-13 | fail→pass | 17,232 | 19,992 | +16% | 1 | 1 | 0% | 2,598 | 3,080 | +19% | 0 | 0 | — |
case-14 | pass→pass | 14,445 | 32,091 | +122% | 1 | 1 | 0% | 2,329 | 5,189 | +123% | 0 | 0 | — |
case-15 | pass→pass | 24,037 | 21,650 | -10% | 1 | 1 | 0% | 3,292 | 3,457 | +5% | 0 | 0 | — |
case-16 | pass→pass | 21,944 | 31,633 | +44% | 1 | 1 | 0% | 3,159 | 5,060 | +60% | 0 | 0 | — |
case-17 | fail→pass | 17,479 | 24,813 | +42% | 1 | 1 | 0% | 2,497 | 4,076 | +63% | 0 | 0 | — |
case-18 | pass→pass | 20,877 | 22,944 | +10% | 1 | 1 | 0% | 2,822 | 3,695 | +31% | 0 | 0 | — |
case-19 | pass→pass | 16,159 | 26,363 | +63% | 1 | 1 | 0% | 2,398 | 4,420 | +84% | 0 | 0 | — |
case-20 | pass→pass | 16,113 | 20,387 | +27% | 1 | 1 | 0% | 2,425 | 3,440 | +42% | 0 | 0 | — |
case-21 | pass→pass | 16,694 | 23,920 | +43% | 1 | 1 | 0% | 2,468 | 3,984 | +61% | 0 | 0 | — |
case-22 | pass→pass | 14,608 | 18,312 | +25% | 1 | 1 | 0% | 2,285 | 3,301 | +44% | 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 +18 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 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.