counterfactual-causal-analysis
SkillDev toolsLets your agent test causal claims by building counterfactual scenarios and finding which factors truly matter.
Available today. Use it from your connected AI after setup.
No other account needed.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the counterfactual-causal-analysis skill
About this skill
Intervene on factors, construct counterfactual worlds, test necessity/sufficiency, and identify load-bearing causal factors.
What this skill tells your AI
The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/counterfactual-causal-analysis/SKILL.md and read by ahel’s review.
Purpose
Intervene on factors, construct counterfactual worlds, test necessity and sufficiency, and identify load-bearing causal factors.
Input contract
required: [causal_claim, variables, intervention_target]
optional: [mechanism_graph, baseline_world, comparison_conditions]
constraints: [intervention and outcome must be measurable or explicitly qualitative]
Execution protocol
Do not perform called SOP operations inline; each loaded SOP owns its contract and thresholds.
- You MUST load skill
extract-causal-structureto extract the causal structure. You MUST load skillidentify-variablesto type its variables. - You MUST load skill
construct-counterfactualto construct the counterfactual. You MUST load skillidentify-load-bearing-factorsto identify load-bearing factors. - You MUST load skill
search-minimal-flipto search the minimal conclusion flip. You MUST load skillevaluate-necessity-sufficiencyto test necessity and sufficiency. If the conclusion should instead be tested by contradiction and counterexample, considerreductio-counterexample-analysisas the next tactic. Deviation: use qualitative counterfactuals when intervention data are unavailable, but mark the inference uncertainty.
Output contract
produces: [causal_structure, counterfactual_map, load_bearing_factors, minimal_flip, necessity_sufficiency_report]
delta_fields: [findings, evidence_updates, hypothesis_updates, uncertainties, decisions]
Thresholds and quality gates
- Every intervention must state changed variables, held-fixed conditions, predicted outcome, and evidence basis.
- Necessity and sufficiency are separate judgments; do not collapse them.
Failure and counterexamples
Do not call a factor necessary when an untested substitute can produce the outcome. Mark counterfactuals underdetermined when held-fixed conditions are unspecified.
Provenance map
- resolved: counterfactual-probing
- resolved: structural-counterfactual
- resolved: necessity-sufficiency
- resolved: systematic-factor-ablation
- resolved: causal-necessity-testing
- resolved: minimal-change-search
Preserved source criteria ledger
| source | source line | kind | source criterion |
|---|---|---|---|
| v4 architecture | node desc | textual | Intervene, construct counterfactual worlds, test necessity/sufficiency, identify load-bearing factors. |
Context checkpoint / Delta notes
Append intervention worlds, held-fixed conditions, factor ranking, flip search, and necessity/sufficiency results.
Signals
- GitHub stars
- 501
- Forks
- 41
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Key
counterfactual-causal-analysis- Source
- github.com/yogsoth-ai/de-anthropocentric-research-engine