extract-concepts

SkillDev tools

Lets your agent pull concepts, definitions, and relations from a source text while keeping each one tied to its exact source wording.

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 extract-concepts skill

About this skill

Extract candidate concepts from evidence with explicit source provenance and precision/recall posture.

What this skill tells your AI

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

Purpose

Extract domain concepts, definitions, and relation candidates from a source while preserving source wording and scope.

Input contract

required: [source, extraction_scope]
optional: [concept_schema, domain_ontology, relation_cues]
constraints: [each concept requires a source span and an uncertainty state]

Procedure

  1. Identify repeated technical terms, defined entities, and salient constructs.
  2. Capture source definitions, aliases, scope qualifiers, and examples.
  3. Link candidate relations only when the source expresses them.
  4. Emit a deduplicated concept set with unresolved meanings retained.

If candidate concepts are identified but still bundle multiple meanings or operations, consider atomize-concept as the next tactic.

Output contract

produces: [concept_records, definition_map, alias_map, relation_candidates, ambiguity_log]
delta_fields: [findings, evidence_updates, uncertainties, open_questions]

Quality gates

  • Concepts are source-linked and distinct from inferred themes.
  • Alias merging preserves conflicting definitions.

Failure and counterexamples

Do not infer a concept solely from keyword frequency or merge terms with different operational meanings.

Provenance map

  • resolved: extract-concepts

Signals

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