ontology

Manage typed knowledge graphs for structured agent memory and cross-skill state sharing.

375|38|Updated Dec 2, 2025
One-click install
npx skills add https://github.com/CraftOS-dev/CraftBot --skill ontology-craftos-dev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ontology
Source: https://github.com/CraftOS-dev/CraftBot/tree/main/skills/ontology
Command: npx skills add https://github.com/CraftOS-dev/CraftBot --skill ontology-craftos-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides a structured way to manage and query information as a typed knowledge graph, enabling robust memory and inter-skill data sharing.

Core Features & Use Cases

  • Entity & Relation Management: Create, query, update, and link entities like Persons, Projects, Tasks, and Events.
  • Schema & Constraints: Define types, properties, and enforce validation rules for data integrity.
  • Use Case: Remember details about a project, link tasks to it, and then query for all open tasks associated with that project.

Quick Start

Use the ontology skill to remember that Alice is a person.

Frequently Asked Questions about ontology

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

FAQPage Schema
How do I structure a typed knowledge graph for agent memory?

A typed knowledge graph manages structured agent memory by enabling entity and relation CRUD, schema definition, and constraint validation. It supports planning multi-step actions as graph transformations and cross-skill state sharing via local JSONL and YAML files.

How do I validate constraints in a knowledge graph schema?

Constraint validation in a knowledge graph schema is enforced by defining specific types, properties, and validation rules. This ensures data integrity when managing entities and relations within local JSONL and YAML files.

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

Sharing state across composable skills is achieved by using a typed knowledge graph that persists entity and relation data via local JSONL and YAML files. This structured approach allows multiple skills to query and update shared memory.

Can I plan multi-step actions as graph transformations?

Yes, you can plan multi-step actions as graph transformations by updating entities and relations within the typed knowledge graph. This models sequential workflows structurally while maintaining schema constraint validation.

How do I query open tasks linked to a project in a knowledge graph?

To query open tasks linked to a project, you create entity types for Projects and Tasks, define a linking relation, and traverse the typed knowledge graph. Schema constraints ensure the queried relational data maintains integrity.