factor-enumeration

$npx mdskill add yogsoth-ai/de-anthropocentric-research-engine/factor-enumeration

Enumerates all key factors, conditions, and assumptions supporting an artifact's conclusion.

  • Identifies dependencies and assumptions in conclusions for deeper analysis.
  • Uses subagent execution to ensure comprehensive and unbiased factor scanning.
  • Analyzes input artifact and optional causal claims to extract relevant factors.
  • Returns structured list of factors with types, importance, and categorization.

SKILL.md

.github/skills/factor-enumerationView on GitHub ↗
---
name: factor-enumeration
description: List all key factors, conditions, and assumptions that support or enable the artifact's conclusion.
execution: subagent
prompt: ./prompt.md
input: artifact (string), causal_claims (list)
used-by: [counterfactual-probing]
---

# Factor Enumeration

Lists all factors that the conclusion depends on — explicit conditions, implicit assumptions, and background requirements.

## Execution

Subagent — spawned via subagent-spawning/spawn-agent.

## Why Subagent

Comprehensive factor enumeration requires systematic scanning without being biased by which factors seem most important.

## Input

- **artifact**: The artifact to analyze
- **causal_claims**: Previously extracted causal claims (optional)

## Output

- **factors**: List of {name, type, explicit/implicit, suspected_importance}
- **factor_count**: Total factors identified
- **categories**: Grouping of factors by type

## Budget

One unit = one enumeration pass per artifact.

More from yogsoth-ai/de-anthropocentric-research-engine