downgrade-equivalence-claim

SkillDev tools

Lets your agent weaken a failed equivalence claim to the strongest still-defensible relation and record what was lost.

Available today. Use it from your connected AI after setup.

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 downgrade-equivalence-claim skill

About this skill

When full equivalence fails, downgrade to the strongest defensible relation (e.g. substructure, homomorphism, shared invariant, analogy) and record exactly what ceased to be preserved.

What this skill tells your AI

The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/downgrade-equivalence-claim/SKILL.md and read by ahel’s review.

Purpose

When full equivalence fails, downgrade to the strongest defensible relation and record what was not preserved.

Input contract

required: [equivalence_claim, preservation_report, counterexamples]
optional: [relation_ladder, scope]
constraints: [each downgrade must name lost structures or operations]

Procedure

  1. List failed preservation obligations.
  2. Test candidate weaker relations.
  3. Select the strongest relation supported and document exclusions.

Output contract

produces: [downgraded_relation, lost_invariants, supporting_mapping, exclusions]
delta_fields: [findings, evidence_updates, uncertainties, decisions]

Quality gates

  • Downgrade cannot silently retain a failed invariant.

Failure and counterexamples

Do not downgrade by vocabulary alone; provide the preserved relation and counterexample boundary.

Provenance map

  • resolved: isomorphism-falsification

Signals

GitHub stars
501
Forks
41
Last commit
Sep 2026
Advanced
Catalog kind
skill
Key
downgrade-equivalence-claim
Source
github.com/yogsoth-ai/de-anthropocentric-research-engine