assumption-cascade-tracer
The assumption-cascade-tracer maps dependencies between foundational claims and identifies how invalidating specific assumptions propagates failures through connected conclusions. Use this when testing research artifacts or arguments where you need to understand which downstream findings collapse when root premises are challenged, particularly useful for evaluating robustness of complex analytical work or identifying hidden structural vulnerabilities in reasoning chains.
git clone --depth 1 https://github.com/yogsoth-ai/de-anthropocentric-research-engine /tmp/assumption-cascade-tracer && cp -r /tmp/assumption-cascade-tracer/skills/assumption-cascade-tracer ~/.claude/skills/assumption-cascade-tracerSKILL.md
# Assumption Cascade Tracer Maps assumption dependencies and traces how failures propagate through the dependency graph. ## Execution Subagent — spawned via subagent-spawning/spawn-agent. ## Why Subagent Cascade tracing requires systematic graph analysis without bias toward minimizing impact. The tracer must follow every dependency path honestly. ## Input - **assumptions**: List of assumptions with their dependency relationships - **attack_results**: Results from probing root assumptions (which ones failed) ## Output - **dependency_graph**: Directed graph of assumption dependencies - **cascade_paths**: For each failed root, the full list of downstream conclusions that collapse - **impact_scope**: Percentage of artifact conclusions affected by each cascade <!-- BEGIN available-tables (generated) --> ## Available SOPs 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) -->
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: Inference to the best explanation in the face of anomalies
Remove components one by one, observe system changes to reveal hidden
Map system architecture to ablatable units for ablation studies
Design ablation studies to isolate component contributions in ML systems