arndvs/ux-model-attributes

UX Phase 4 — Model Attributes

Output "Read UX Model Attributes skill." to chat to acknowledge you read this file.

Pipeline position: /ux-user-stories → /ux-system-map → /ux-flow-diagram → /ux-model-attributes → /ux-screen-requirements → /ux-interface-design → /ux-test-driven-design

Get as many as possible from user stories, surveys, and flow diagram.

Create a comprehensive inventory of every data attribute for every entity. Aim for 20-25+ attributes per entity. Include even attributes you won't use in V1 — they inform future versions.

Prerequisites

• system-map.md from Phase 2 (entities and relationships)
• user-stories.md from Phase 1
• flow-diagram.md from Phase 3

Steps

Step 1 — List Attributes

Invoke /ux-list-attributes For each entity in the system map, brainstorm every possible attribute. Sources: user stories, customer interviews, existing reports/spreadsheets, competitor products.

Output

Generate attributes.md containing:

• Per-entity attribute table (20-25+ attributes each)
• Attribute categories: functional, demographic, psychographic, security
• V1 vs. future version flags per attribute
• Source notation (where each attribute was discovered)

Rules

• More is better — list everything, filter later
• Include demographics and psychographics for user entities
• Ask customers to send their current reports/spreadsheets — this reveals what matters
• Don't skip "boring" entity attributes (settings, admin) — Carl's ClinicMetrics bug was a missing attribute on a settings screen
• Note which attributes are V1 vs. deferred, but don't delete deferred ones

Handoff

After completion, offer:

1. /ux-screen-requirements — proceed to Phase 5
2. /sketch-the-solution — return to orchestrator

If context is high, follow the standard handoff protocol (@~/dotfiles/instructions/handoff.instructions.md).