ontology

Store and validate typed knowledge graphs for agent memory.

8|1|Updated Apr 13, 2026
One-click install
npx skills add https://github.com/Tugoukezhang/workbuddy-skills --skill ontology-tugoukezhang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ontology
Source: https://github.com/Tugoukezhang/workbuddy-skills/tree/main/skills/ontology
Command: npx skills add https://github.com/Tugoukezhang/workbuddy-skills --skill ontology-tugoukezhang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve?

The Ontology skill provides a typed knowledge graph to represent agent memory and enable composable skills, empowering structured entities, relations, and graph-driven planning across tasks.

Core Features & Use Cases

  • Type-safe entities and relations with constraints
  • Graph transformations for planning multi-step work
  • Cross-skill data sharing and memory persistence across skills

Use cases include creating Person/Project/Task graphs, querying dependencies, and planning workflows by mutating the ontology graph.

Quick Start

Create a Person named Alice and relate it to a Project in the ontology graph.

Frequently Asked Questions about ontology

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

FAQPage Schema
How do I store structured agent memory as a typed knowledge graph?

You store structured agent memory as a typed knowledge graph by defining type-safe entities and relations with constraints. This approach enables schema-driven validation and persists memory across tasks using an append-only graph storage format.

What is the best way to share state across multiple agent skills?

The best way to share state across multiple agent skills is using a typed knowledge graph for cross-skill data sharing. By mutating the ontology graph, you enable graph-driven planning and maintain memory persistence across different workflows.

How do I plan multi-step workflows using graph transformations?

You plan multi-step workflows using graph transformations by creating entities like Person, Project, or Task and linking their relations. Mutating the ontology graph allows you to map dependencies and sequence actions across your tasks.

Can I validate entities and relations with schema-driven constraints?

Yes, you can validate entities and relations with schema-driven constraints. The typed knowledge graph enforces type-safe structures, ensuring that cross-skill state and agent memory conform to defined schemas during CRUD operations.

Do I need pyyaml to manage an append-only graph database for agent memory?

Yes, you need pyyaml to manage the append-only graph database for agent memory. The dependency supports the schema-driven validation required to store and manipulate structured graph state at the designated file path.

How do I query dependencies between Person, Project, and Task entities?

You query dependencies between Person, Project, and Task entities by creating the objects and linking their relations within the typed knowledge graph. CLI CRUD operations allow you to retrieve and traverse these structured connections for planning.