pattern-stack-architect

Pattern Stack's Atomic Architecture organizes backend code into four layers—Atoms, Features, Molecules, and Organisms—each with distinct responsibilities. This layered approach enforces separation of concerns, prevents anti-patterns, and ensures consistent pattern composition across SQLAlchemy models, services, and domain logic.

Use Field() for cleaner type inference, built-in validation, and Pattern Stack compliance. Field() eliminates mapped_column verbosity, provides type hints that flow through your models, and integrates with Atoms, Features, Molecules, and Organisms for consistent domain modeling.

Pattern Stack supports multiple inheritance to compose core patterns—Catalog, Actor, Event, Temporal, and Hierarchical—into robust domain models. Combine patterns by inheriting from multiple pattern classes; the Skill guides correct composition to avoid conflicts and ensure layer boundaries remain intact.

Atoms are foundational building blocks; Features add business logic; Molecules combine Features; Organisms orchestrate complete workflows. Each layer has explicit boundaries; the Skill enforces these constraints, prevents cross-layer violations, and ensures correct implementation of services, entities, and APIs.

The Skill reviews your code against Pattern Stack's layer structure and pattern composition rules, catching violations early. It validates Field() usage, pattern inheritance chains, and API/service design to prevent anti-patterns and keep your architecture aligned with Atomic Architecture principles.

Yes. Pattern Stack patterns apply to SQLAlchemy models, services, entities, and domain workflows. The Skill guides you through using patterns for data layer design and orchestrating those patterns into higher-level Molecules and Organisms for business logic and API design.