assumption-excavation
Assumption Excavation systematically identifies hidden beliefs underlying decisions, then adversarially challenges each one and maps which assumptions would change the conclusion if proven wrong. Use this skill when evaluating major decisions, strategies, or arguments where unstated premises might be fragile or where understanding load-bearing beliefs is essential for risk assessment and mitigation planning.
git clone --depth 1 https://github.com/yogsoth-ai/de-anthropocentric-research-engine /tmp/assumption-excavation && cp -r /tmp/assumption-excavation/skills/assumption-excavation ~/.claude/skills/assumption-excavationSKILL.md
# Assumption Excavation A three-phase tactic that surfaces hidden assumptions, challenges each one adversarially, and maps which assumptions are load-bearing for the conclusion. Decisions often rest on unstated beliefs — this tactic makes them explicit and tests their strength. ## Stages 1. **Assumption Extraction** — Systematically surface all assumptions underlying the decision, with confidence levels 2. **Assumption Challenge** — For each assumption, construct the strongest counter-argument and identify alternatives 3. **Conclusion Sensitivity** — Map which assumptions, if wrong, would change the conclusion ## Available SOPs | SOP | Phase | Purpose | |-----|-------|---------| | assumption-extraction | Extract | Surface hidden assumptions with confidence | | assumption-challenge | Challenge | Attack each assumption adversarially | | conclusion-sensitivity | Sensitivity | Map load-bearing assumptions | ## Execution Guidance - Extract minimum 5 assumptions per decision - Challenge ALL assumptions, not just obvious ones - Confidence levels: HIGH (>80%), MEDIUM (50-80%), LOW (<50%) - Critical assumption = conclusion changes if assumption is wrong - Focus mitigation efforts on critical + low-confidence assumptions ## Minimum Yield - >= 5 assumptions extracted with confidence levels - Challenge argument for each assumption - Alternative assumption for each (what if the opposite is true?) - Sensitivity map showing which assumptions are critical - List of critical assumptions requiring mitigation
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