semantic-relations

Classify semantic relation types and properties in knowledge graph edges.

1|Updated May 6, 2026
One-click install
npx skills add https://github.com/jacob-balslev/skill-graph --skill semantic-relations
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: semantic-relations
Source: https://github.com/jacob-balslev/skill-graph/tree/main/marketplace/skills/semantic-relations
Command: npx skills add https://github.com/jacob-balslev/skill-graph --skill semantic-relations

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Semantic-relations helps you avoid confusing or mislabeling the meaning of connections between concepts, so your knowledge graph edges support correct hierarchy, disambiguation, and boundary decisions.

Core Features & Use Cases

  • Relation-type coverage: Distinguishes taxonomic (IS-A), mereological (PART-OF), associative/lexical (synonymy, antonymy, polysemy, homonymy, metonymy), and thematic (agent, patient, instrument, cause, goal, etc.) relations.
  • Relation properties discipline: Applies symmetry, asymmetry, transitivity, reflexivity, and irreflexivity to prevent direction and closure mistakes.
  • Verification & anti-pattern detection: Uses a checklist to catch vague related_to edges, circular IS-A, IS-A vs PART-OF conflation, synonym sprawl, untyped polysemy, and property-free relation definitions.
  • Boundary routing: Clarifies when the task belongs to adjacent skills (e.g., linguistics, ontology-modeling, taxonomy-design, entity-relationship-modeling) versus relation typing.

Quick Start

Ask the agent to audit a set of proposed graph edges and classify each edge into the correct semantic relation type, including which relation properties apply and which likely anti-patterns it matches.

Frequently Asked Questions about semantic-relations

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent incorrect semantic labeling of knowledge graph edges?

To prevent incorrect semantic labeling of knowledge graph edges, apply a typed-relation taxonomy with substitution tests to classify IS-A and PART-OF claims, verifying symmetry, asymmetry, transitivity, and reflexivity expectations explicitly.

What is the difference between IS-A and PART-OF hierarchy claims in a concept map?

IS-A and PART-OF hierarchy claims differ fundamentally: IS-A represents taxonomic subsumption validated by substitution tests, while PART-OF represents mereological part-whole semantics requiring distinct validation checks to prevent conflation.

How do I audit knowledge graph relations for anti-patterns like circular IS-A or synonym sprawl?

Audit knowledge graph relations for anti-patterns by applying a verification checklist that detects vague related_to edges, circular IS-A structures, IS-A vs PART-OF conflation, synonym sprawl, and untyped polysemy.

Does this approach to relation typing commit to a specific formal ontology or database implementation?

No, this approach to relation typing does not commit to a specific formal ontology or database implementation; it focuses purely on classifying edge types and relation properties without making implementation commitments.

When should I route semantic relation tasks to adjacent skills like taxonomy-design or ontology-modeling?

Route semantic relation tasks to adjacent skills like taxonomy-design or ontology-modeling when the work shifts from classifying edge types and relation properties to broader structural design, linguistic analysis, or entity-relationship modeling.

How do I validate thematic roles and lexical relations like polysemy and homonymy in a knowledge graph?

Validate thematic roles and lexical relations like polysemy and homonymy by applying a typed-relation taxonomy that distinguishes agent, patient, instrument, cause, and goal connections alongside synonym, antonym, and metonymy distinctions.