What problem does it solve? Domain terminology drifts as a project grows: the same concept gets called 'account', 'customer', and 'user' in different places, and the reasoning behind key architectural choices gets lost. This Skill actively sharpens a project's domain model during design discussions and records the results as a living glossary and decision log. ## Core Features & Use Cases - Glossary Management: Creates and updates CONTEXT.md files with opinionated canonical terms, tight definitions, and explicit 'Avoid' synonyms, supporting both single-context and multi-context repos via CONTEXT-MAP.md. - Architecture Decision Records: Writes minimal ADRs in docs/adr/ with sequential numbering, offered only when a decision is hard to reverse, surprising without context, and the result of a real trade-off. - Active Modeling Discipline: Challenges terms that conflict with the glossary, sharpens fuzzy language, stress-tests domain relationships with edge-case scenarios, and cross-references claims against the actual code. - Use Case: While designing a billing feature, the user says 'the account gets charged'. The Skill flags that CONTEXT.md defines 'Customer' and 'Invoice' but not 'account', resolves the term inline, updates the glossary, and offers an ADR when the team decides Billing communicates with Ordering via domain events rather than synchronous HTTP. ## Quick Start Ask the AI to help define the domain language for your project and record the key architectural decisions as ADRs.
