abstract-structure

SkillDev tools

Lets your agent strip away surface details from a problem to reveal the underlying structure that transfers to other domains.

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 abstract-structure skill

About this skill

Remove domain surface details to expose transferable relational/mechanistic structure at a chosen abstraction level.

What this skill tells your AI

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

Purpose

Strip domain surface detail and retain transferable relations or mechanisms at a declared abstraction level.

Input contract

required: [source_case, abstraction_level]
optional: [relation_schema, invariants]
constraints: [preserve causal or functional roles while removing incidental labels]

Procedure

  1. List entities, actions, constraints, and outcomes in the source case.
  2. Replace domain labels with role and relation types at the requested level.
  3. Record invariants and details intentionally discarded.

Output contract

produces: [abstract_structure, retained_invariants, discarded_details]
delta_fields: [findings, hypothesis_updates, uncertainties]

Quality gates

  • The abstraction level is named; every retained relation has source support; no causal relation is introduced by analogy.

Failure and counterexamples

Return an abstraction gap when role mapping is ambiguous or when removing a detail changes the mechanism.

Provenance map

  • creative-ideation/abstraction-extraction: resolved.
  • creative-ideation/abstraction-ladder: resolved.
  • creative-ideation/generic-space-extraction: resolved.

Signals

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