Conciseness

The body is dense and assumes SwiftUI competence — pattern entries like "GeometryReader has no intrinsic size, so a stack gives it the space left after fixed-size siblings" convey non-obvious facts in one line. It stays at anchor 4 rather than 5 because the five invocation examples, the 'why it matters' column in Phase 3, and some repeated iOS-26/27 resizing rationale across sections could be trimmed without losing guidance.

Actionability

Every detection pattern ships an exact grep regex (e.g. `\.frame\(width:\s*\d{3,}`, with explicit caveats about what the regex over-matches and how to Read-verify), fixes include copy-paste code (`.containerRelativeFrame(.horizontal) { w, _ in w * 0.5 }`), and the output format is a complete fill-in markdown template with per-issue fields.

Workflow Clarity

Five phases are clearly sequenced with an explicit checkpoint ("Present this map in the output before proceeding") and validation throughout: "use Read to verify the surrounding context before reporting", a dedicated False Positives section, and output limits for large findings. The operation is read-only auditing, so the destructive-operation cap does not apply.

Progressive Disclosure

The body is well-sectioned but entirely monolithic — no references/, scripts/, or assets/ bundle files exist, and all 12 detection patterns, the Phase 3/4 tables, and scoring rubric are inlined in one 300+ line file. The only pointers ("axiom-swiftui skills/toolbars.md, skills/iphone-duo.md") target an external skill not in this bundle. This matches anchor 3 ('content that should be separate is inline') — structure is good, but the per-pattern detail could be split into a references file so the main file stays an overview.

Specificity

The description names the domain ("SwiftUI layout review", "adaptive layout issues") but contains no capability statements at all — there is not a single action verb describing what the skill does. It fits anchor 2 ('names the domain but actions are minimal or generic') rather than 1, since the language is domain-specific rather than purely abstract, but cannot score higher because zero actions are listed.

Completeness

This is a pure trigger clause: "Use when the user mentions..." — the 'when' is explicit but the 'what' is entirely absent, matching anchor 2 ('only when is present without what', e.g. "Use when working with documents"). It cannot score 3 because anchor 3 requires a clear 'what' even with a weak 'when'.

Trigger Term Quality

"SwiftUI layout review", "adaptive layout issues", "GeometryReader problems", and "multi-device layout checking" are phrases users would naturally say. It falls short of anchor 5 because common synonyms are missing — users say "iPad", "Split View", "rotation", or "responsive" when describing these problems, and none appear.

Distinctiveness Conflict Risk

"GeometryReader problems" and "adaptive layout issues" are distinct triggers establishing a clear niche. It stays at anchor 4 rather than 5 because "SwiftUI layout review" could catch general SwiftUI code-review requests, and the body itself acknowledges overlap with sibling auditors (resize-auditor, swiftui-performance-analyzer).