What problem does it solve? Data-structure designs such as DDL, schemas, ER diagrams, and CRUD notes often hide unresolved behavior that an AI would otherwise fill in with unsupported guesses. This Skill mechanically derives requirement confirmation gates from the design before implementation starts, so ambiguous cases like null handling, not-found behavior, and multiplicities are confirmed by humans instead of being silently assumed. ## Core Features & Use Cases - Mechanical DCD Derivation: Enumerate design elements by type, constraint, relation, operation, and business pattern, then apply the design constraint catalog to assign DCD-prefixed items. - Separation of Concerns: Split requirement confirmations from design-time decisions, and expand combination matrices only when structural axes actually exist. - Explicit Unresolved Items: Emit anything unsupported as unresolved rather than inventing defaults, and write results to a dated Markdown file under work/constraint_derivation/. - Use Case: Given a new database schema with nullable columns and one-to-many relations, run this Skill to produce a derivation basis table and a grouped list of questions the product owner must confirm before coding begins. ## Quick Start Ask the AI to derive design constraints and requirement confirmation items from your DDL or ER design document and write the result to a Markdown file.
