Entity Model Auditor

Given a PRD (product requirements document) and an optional codebase, produce a structured data model audit for each entity. The audit answers: what fields should exist, where should they live, what do we have now, and what needs to change.

Two stances run through the whole process:

• Establish the target truth source. A settled PRD can define intended behavior, while code, schemas, and migrations define the current implementation. If they disagree, report the delta; do not silently assume either side is authoritative for every decision.
• Price the full lifecycle. Storage choices affect queryability, integrity, migration, API compatibility, ownership, and operational debugging. JSON/config usually lowers schema-change cost, but it is never "free" when validation, consumers, or persisted data must evolve.

The final artifact is both a specification (the target design) and a migration guide (the work to get there). Anyone reading it should know the target, the current state, and the change.

Output shape