estimate-effective-evidence-count
SkillDev toolsLets your agent check how many pieces of evidence are truly independent by discounting shared sources and correlations.
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 estimate-effective-evidence-count skill
About this skill
Estimate an effective number of independent evidence/reasoning paths after discounting dependence/correlation; report nominal N, dependence structure, and N_eff.
What this skill tells your AI
The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/estimate-effective-evidence-count/SKILL.md and read by ahel’s review.
Purpose
Estimate independent evidence count after discounting dependence and correlation.
Input contract
required: [evidence_paths, dependence_structure]
optional: [correlation_estimates, nominal_count]
constraints: [dependence assumptions and uncertainty must be explicit]
Procedure
- Count nominal paths and trace shared dependencies.
- Assign dependence classes or correlations.
- Estimate N_eff and sensitivity to dependence assumptions.
Output contract
produces: [nominal_count, dependence_map, effective_count, sensitivity_range]
delta_fields: [findings, evidence_updates, uncertainties, decisions]
Quality gates
- Report N and N_eff separately; include a range when dependence is uncertain.
Failure and counterexamples
Do not inflate N_eff by counting repeated sources or shared models as independent.
Provenance map
- resolved: independent-convergence-audit
Signals
- GitHub stars
- 501
- Forks
- 41
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Key
estimate-effective-evidence-count- Source
- github.com/yogsoth-ai/de-anthropocentric-research-engine