Reverse-Engineer ArtisanShop into AI Unified Process Artifacts

Background

ArtisanShop is a Java/Spring Boot marketplace platform where independent artisans can list handmade products, buyers can browse and purchase them, and administrators manage the platform. The codebase was written two years ago and has grown significantly — it now spans multiple feature areas: authentication, user profiles, a product catalog, shopping and order fulfillment, notifications, and an admin back office with reporting.

The engineering lead wants to onboard new developers using the AI Unified Process methodology, starting with a full set of AI Unified Process artifacts derived from the existing codebase. The goal is to have a use case diagram, individual use case specifications, and an entity model that a developer joining the team could read instead of spelunking through the source code.

The codebase is in the inputs/ directory. It contains a pom.xml, Spring Security configuration, controller classes spread across several packages, JPA entity classes, and Flyway SQL migrations under src/main/resources/db/migration/.

Your Task

Each artifact must conform exactly to the standard AI Unified Process document format — diagram layout, spec structure, and entity model attribute tables using the AI Unified Process data-type and validation-rule vocabulary (a fixed list of seven data types — never raw SQL types and never ad-hoc descriptive types). Documents that deviate from the standard format will be rejected during review.

Reverse-engineer the ArtisanShop codebase into the three AI Unified Process artifacts:

1. docs/use_cases.puml — PlantUML use case diagram listing all actors and use cases
2. docs/use_cases/ — one specification file per use case (e.g. UC-001-place-order.md)
3. docs/entity_model.md — entity model with Mermaid ER diagram and attribute tables

Because the codebase covers many features, document your approach in docs/PLAN.md as you work. The plan file must record:

• The full list of controller files you found in your initial scan of the codebase
• How you grouped those controller files into feature clusters before writing specs
• Which database tables you decided to exclude from the entity model, and why

Produce the docs/PLAN.md file first, then generate the three AI Unified Process artifacts.

Output Specification

All output files go under docs/:

• docs/PLAN.md — clustering plan (controller list, feature groups, excluded tables)
• docs/use_cases.puml — PlantUML diagram
• docs/use_cases/UC-XXX-short-name.md — one file per use case (kebab-case filenames)
• docs/entity_model.md — entity model

Do not modify any file under inputs/.