Dense, reference-style content with tables and tight bullets; it does not explain concepts Claude already knows (no 'what is A/B testing' padding), so most tokens earn their place even at ~220 lines.

Concrete, copy-paste-ready guidance throughout — numbered App Store Connect steps, a hypothesis template with worked examples, and specific rules ('Change ONE thing per test', '90% confidence minimum', sample-size bands) rather than abstract direction.

The Test Design Framework is sequenced as Steps 1–5 with explicit checkpoints ('Monitor but don't stop early', statistical-significance checks in Step 5) and an interpretation checklist, matching the clear-sequence anchor.

The body is well-sectioned but monolithic at ~220 lines, with the Common Test Ideas tables and Output Format templates kept inline rather than split into reference files; it sits above the 'monolithic wall' anchor but below the 'appropriately split, one-level-deep references' level.

Names a concrete domain and action ('A/B test App Store product page elements to improve conversion rate') with specific testable elements, matching the 'lists multiple specific concrete actions' anchor rather than the vague 'names domain and some actions' level below.

Explicitly answers both what ('A/B test App Store product page elements to improve conversion rate') and when ('Also use when the user mentions...'), satisfying the 'clearly answers both what AND when' anchor.

Provides strong coverage of natural terms users would say — 'A/B test', 'test my screenshots', 'test my icon', 'conversion rate optimization', 'CPP', 'custom product pages' — matching the good-coverage anchor.

Clear App Store PPO/CPP niche with distinct triggers and explicit pointers to sibling skills (screenshot-optimization, metadata-optimization), making it unlikely to trigger for the wrong skill.