ontology

A typed knowledge graph represents entities, their properties, and inter-entity relations with enforced constraints. It ensures verifiable mutations and consistent reasoning across skills by applying typed vocabularies to memory, planning, and data access operations.

You need a typed knowledge graph when modeling complex entities like Projects and Tasks across shared skills. It provides graph traversals and dependency queries to maintain consistent reasoning, verifiable mutations, and structured data access.

Initialize ontology storage at memory/ontology, then use CLI-based operations to create, relate, and validate entities such as Person, Project, Task, Event, and Document with graph traversals and dependency tracking.

Yes, you can define custom entity types and relations using an optional YAML schema for type definitions. This allows you to enforce specific constraints and validate entities within your local memory graph.

The best way to validate inter-entity relations and track dependencies is using a typed knowledge graph with enforced constraints. It applies CLI-based operations to ensure verifiable mutations and consistent reasoning across entities.

Yes, the typed knowledge graph supports graph traversals and dependency queries across multiple skills. It shares state and plans multi-step transformations by modeling workflows as graph operations against a local memory graph.