ontocubes 0.1.0

Universal ontological cube framework - represent any entity as a 5-layer cube (temporal, typology, ontology, causality, individuality)

ontocubes

Universal ontological cube framework for representing any entity as a 5-layer cube.

Concept

OntoCubes provides a unified model where any entity in the world can be represented as a cube with 5 layers:

Layer	Purpose	Example
Temporal	Identity & versioning	id, version, timestamp, name
Typology	Type hierarchy	TEXT.PROMPT.LLM
Ontology	Semantic classification	{"creative/storytelling": 0.95}
Causality	Parent/ancestry chain	Reference to parent cube
Individuality	Actual data (delta from type)	{prompts: {system: "...", user: "..."}}

Key Features

• Self-referential: Same schema in → same schema out
• Recursive resolution: {{ref}} syntax resolves cubes recursively
• Type-aware processors: Transform data between types (TEXT → PROMPT)
• Schema validation: Root cubes define schemas for their type hierarchy
• SQLite persistence: Fast lookup with YAML as source of truth

Status

🚧 Under Development - API may change

Installation

pip install ontocubes

Quick Example

from ontocubes import Storage, Resolver

# Store a cube
storage = Storage.get()
storage.store({
    "name": "greeting",
    "typology": "TEXT",
    "individuality": {"content": "Hello, World!"}
})

# Resolve with refs
storage.store({
    "name": "prompt",
    "typology": "TEXT.PROMPT.LLM",
    "individuality": {
        "prompts": {
            "system": "{{greeting}}",  # Resolves to "Hello, World!"
            "user": ""
        }
    }
})

resolver = Resolver()
result = resolver.resolve("prompt")
print(result["individuality_resolved"]["prompts"]["system"])
# Output: Hello, World!

License

MIT