In the Skill's Own Terms
Connects to a real iPhone via gstack-ios-qa-daemon, screenshots every screen, and evaluates visual design against Apple HIG, DESIGN.md, and best practices. Scores ten dimensions (typography hierarchy, spacing rhythm, color hierarchy, touch targets, loading/empty/error states, accessibility, animation discipline, iOS idiom alignment, information density, and AI-slop check) from 0-10, using 'what would make it a 10' framing.
Mirrors /plan-design-review scoring rubric ported to iOS idioms. Produces a markdown report with screenshots, scores per screen, and biggest-leverage fix suggestions. For plan-stage design review (before implementation), use /plan-design-review instead. For live web visual audits, use /design-review.
What it produces
- Markdown report with screenshots inline, scores per screen (0-10 for each dimension), and biggest-leverage fix suggestions per dimension
How It Works
- 01Connect to Running Daemon
Uses the running gstack-ios-qa-daemon. If no daemon is running, spawn one via the same flow as /ios-qa (Phase 0-2).
- 02Capture Screens
For each major screen (driven from a user-provided screen list or auto-discovered via the accessibility tree), retrieves screenshot and elements via GET /screenshot and GET /elements.
- 03Apply ten-dimension Rubric
Scores typography hierarchy, spacing rhythm, color hierarchy, touch targets, loading/empty/error states, accessibility, animation discipline, iOS idiom alignment, information density, and AI-slop check from 0-10, explaining what would push each to 10.
- 04Flag Low Scores
Uses AskUserQuestion for any score below 7, presenting the issue with recommended fix and tradeoff so the user can decide whether to address it.
- 05Write Report
Produces a markdown report with screenshots inline, scores per screen, and biggest-leverage fix suggestions per dimension, written to ~/.gstack/projects/<slug>/ios-design-review-<date>.md.