alternative-analysis
Alternative-analysis generates competing interpretations and scenarios from the same evidence to counteract cognitive bias and single-explanation anchoring. Use this skill when evaluating threats, analyzing complex systems, or challenging dominant narratives by systematically testing multiple hypotheses against available evidence through what-if analysis, four-lens examination, and discriminating diagnostic tests.
git clone --depth 1 https://github.com/yogsoth-ai/de-anthropocentric-research-engine /tmp/alternative-analysis && cp -r /tmp/alternative-analysis/skills/alternative-analysis ~/.claude/skills/alternative-analysisSKILL.md
# Alternative Analysis Strategy
CIA SAT alternative analysis: generate multiple competing interpretations to prevent anchoring on a single explanation.
## Method
1. **alternative-futures** builds 2-4 divergent scenarios from the same evidence base
2. What-If Analysis: systematically vary key variables to explore outcome sensitivity
3. Four Ways of Seeing: examine artifact through lenses of opportunity, risk, structure, and agency
4. **probe-execution** tests each alternative against available evidence
5. Diagnostic indicators identified that would discriminate between alternatives
6. **finding-aggregation** compares explanatory power across alternatives
## Budget Table
| Parameter | S | M | L |
|---|---|---|---|
| Attack vectors | 5 | 12 | 20 |
| Probing rounds | 3 | 6 | 10 |
| Personas | 2 | 4 | 6 |
| Assumption checks | 5 | 10 | 20 |
## Orchestration
```
threat-surface-mapping → [identify variable dimensions]
→ alternative-futures (generate 2-4 scenarios)
→ [for each alternative]:
attack-vector-generation (find discriminating tests)
→ probe-execution (test alternative)
→ finding-aggregation → attack-resilience-scoring
```
## Subagents
- threat-surface-mapping (dimension identification)
- alternative-futures (scenario generation)
- attack-vector-generation (discriminating test design)
- probe-execution (alternative testing)
- finding-aggregation (comparative synthesis)Experiment-specific - summarize the DARE executor's research design into a clean research_result report, forced to write back into the spec file produced by formated-specs.
Experiment-specific - replaces writing-specs, emits DARE's 4-layer call plan as a clean research_graph schema. Last step forces load formated-result.
loss-1 judge - read a sample's full dialogue and decide whether the user simulator semantically enacted its Policy Card. check-blind.
loss-2 judge - pairwise quality comparison across the n rungs within one topic; decide monotonicity and endpoint separation. check-blind, D1-D5 only.
Strategy: 面对异常的最佳解释推理
Remove components one by one, observe system changes to reveal hidden dependencies and generate ideas from structural gaps.
Map system architecture to ablatable units for ablation studies
Design ablation studies to isolate component contributions in ML systems