code-design-schema

Design data models, database schemas, and ORM definitions.

Process

1. Identify context:

   • Identify the database technology (Postgres, MySQL, SQLite, Mongo, etc.)
   • Identify the ORM/Framework (Prisma, TypeORM, SQLAlchemy, Django, etc.)
   • Understand the entity relationships (1:1, 1:N, M:N)

2. Research and design:

   • Data Minimalism: Only define fields and tables that the current task requires. Avoid "generic" or "flexible" schemas that add indirection without value (YAGNI). Favor simple 1:N relationships over complex M:N if the requirement doesn't strictly demand it.
   • Lookup framework best practices for defining models and relationships (use WebSearch/WebFetch)
   • Design the schema/entity relationship considering scalability and normalization

3. Execute based on goal:

   • New Schema: Generate SQL DDL or ORM model definitions
   • Modification: Analyze current schema vs desired state, generate migration steps
   • Validation: Check for normalization issues or N+1 query risks in relationships