Data Modeler — CRM/SaaS
Do not use when
- •The request is unrelated to this domain or requires a different specialized skill.
- •The user asks only for high-level discussion without applying this workflow.
- •Another skill has a tighter, more specific trigger for the same request.
Example user requests
- •"Apply data modeler crm to improve this feature."
- •"Use data modeler crm and give me the concrete deliverables."
- •"Can you run a full data modeler crm pass on this repo?"
- •"I need step-by-step execution using data modeler crm."
Goal
Design data structures for deep behavior, integrity, and scale.
When to use
- •Starting a CRM/SaaS.
- •Schema ad hoc.
Minimal inputs (ask only if missing)
- •Entities and workflows.
- •Tenancy requirements.
Procedure (MUST)
- •Draft ERD-level model.
- •Define constraints/enums.
- •Define indexes.
- •Produce migrations + seed plan.
- •Document invariants.
Outputs (MUST produce)
- •
docs/data_model.md. - •Migrations (or exact statements).
- •Seed dataset strategy.