Lean and information-dense — tables, exact specs (icon sizes, 20–40% lift, p<0.05, 1000+ impressions), and templates rather than prose. It avoids explaining concepts Claude already knows; the brief "Why" framing supplies actionable domain context, not padding.

Highly concrete guidance: exact UI navigation paths ("App Store Connect → Your App → Product Page Optimization → Create Test"), required size tables, a test-variable matrix, a 1–10 audit checklist, and a copy-ready designer brief template.

Multi-step A/B test flows are numbered with explicit checkpoints ("Run for minimum 7 days or until statistical significance", significance threshold p<0.05, minimum sample), and the icon audit is a structured checklist — clear sequences with verification built in.

Well-organized into clearly headed sections with a "Related Skills" block providing one-level navigation to sibling skills; no nested/deep references and no monolithic wall of inline API detail. The inline guidance is appropriately overview-sized and easy to navigate.

Lists multiple concrete actions — "design, test, or improve their app icon to increase tap-through rate and conversions" — covering design, testing, and improvement rather than vague language. Voice is third person ("the user wants", "the user mentions"), so no voice penalty applies.

Explicitly answers both what (design/test/improve app icons for TTR and conversions in App Store search and browse) and when ("Use when the user mentions..." with a clear trigger list), satisfying the full what+when requirement.

Strong coverage of natural phrases users would say: "app icon", "icon design", "icon A/B test", "icon variants", "tap-through rate", "icon conversion", "icon refresh", plus "what makes a good app icon".

Clear niche (app icon optimization) with explicit disambiguation pointers — "For screenshot optimization, see screenshot-optimization" and "For full listing A/B tests, see ab-test-store-listing" — making conflicts with sibling skills unlikely.