What problem does it solve? When researching a competitor or reference product, teams often guess at the underlying data model from surface impressions. This Skill provides a disciplined method for reconstructing the entities, fields, relationships, and enforced rules of an existing product, with every claim evidenced by the screen, export, or endpoint it was read from. ## Core Features & Use Cases - Entity and field recovery: Identify the nouns a product is built on, each field's name, type, required status, constraints, and the label the product shows for it. - Evidence-first sourcing: Prefer documented models (API references, schema endpoints, export files, import templates) over interface inference, and mark every line as documented or inferred. - Relationship and rule mapping: Capture cardinality, cascade behavior on deletion, validation, uniqueness, ordering, permissions, and derived-versus-stored fields. - Use Case: Researching Folk CRM's person record — walk the product, pull the CSV import template, capture the forms with ui-capture, and produce a per-entity field table with locators for every claim. ## Quick Start Ask the agent to research the schema domain of a specific product, for example: scrape the schema from Folk CRM and document the person entity's fields, relationships, and enforced rules.
