claim-refinement

$npx mdskill add yogsoth-ai/de-anthropocentric-research-engine/claim-refinement

Refines claims to survive counterexamples while maximizing explanatory power

  • Solves the problem of claims being invalidated by counterexamples
  • Uses logical analysis and lemma incorporation techniques
  • Evaluates counterexamples to narrow scope or split claims
  • Returns a revised claim with preserved explanatory power
SKILL.md
.github/skills/claim-refinementView on GitHub ↗
---
name: claim-refinement
description: "Propose a refined claim that survives counterexamples while preserving maximum explanatory power (Lakatos lemma-incorporation)."
execution: subagent
prompt: ./prompt.md
used-by: [adversarial-stress-testing]
---

# Claim Refinement

Subagent that refines claims to survive counterexamples using lemma incorporation, scope narrowing, or claim splitting.
More from yogsoth-ai/de-anthropocentric-research-engine