What problem does it solve? Domain models degrade when the same term means different things to different actors, types accumulate boolean flags and unrelated properties, and specification tables hide cross-cutting rules. This Skill provides a consistent workflow to detect modeling problems, decide between creating or restructuring models, build types that represent domain concepts, and validate the result. ## Core Features & Use Cases - Detection signals: Identifies context boundary signs (language games), ubiquitous language decay (glossary-only lexicon, missing verbs, impossible states), and restructuring smells (fat models, boolean state flags, stamp coupling, shotgun parsing). - Modeling techniques: Covers context boundary discovery by actor purpose, hidden concept extraction from if-statements into Policies/Specifications, type granularity design, specification orthogonality checks, aggregate boundary heuristics (lock contention, cascade delete tests), and the DDD Trilemma for domain service purity. - Validation outputs: Produces mermaid classDiagram domain model diagrams, object diagrams with concrete instances, and NotebookLM-based literature review against DDD and CQRS/ES notebooks. - Use Case: When adding a feature where "order" means different things to sales, shipping, and accounting, use this Skill to split contexts, model states as discriminated unions instead of flags, and verify aggregate boundaries before implementation. ## Quick Start Ask the AI to apply the domain-modeling skill to analyze whether a new business concept needs a new model or a restructure of an existing type.
