Conciseness

The body is lean and assumes Claude's competence (no explanations of known concepts), but the Pattern Selection table and Escalation Signals sections partially overlap — both state when to choose each pattern — and the 9-line Contents list could be trimmed. This fits anchor 4 (efficient with minor instances that could be tightened) rather than anchor 5.

Actionability

A complete, copy-paste-ready Swift example for the MV default (injected TripClient, @Observable store, .task wiring) covers the common case, and escalation/migration guidance is concrete ('Move one state transition or dependency at a time'). Not anchor 4: there are no gaps — the other patterns' structures are deliberately and explicitly deferred to the reference file.

Workflow Clarity

Decision Workflow and Migration are numbered sequences with explicit validation checkpoints and a feedback loop ('If behavior changes, restore the fixture, fix the smallest boundary, and rerun before migrating another slice'), plus a Review Checklist. Matches anchor 5's explicit validation, error-recovery loops, and checklists.

Progressive Disclosure

The body is a well-sectioned overview (with TOC) that keeps only decision guidance inline and defers detailed pattern structures to a real, one-level-deep references/architecture-patterns.md, clearly signaled both inline ('Load [Architecture Pattern Recipes]...') and in the References section. Matches anchor 5.

Specificity

'Choose, review, or migrate Apple-platform architecture and module boundaries' names three concrete actions plus the domain. It falls short of anchor 5's comprehensive coverage because sub-capabilities (test seams, dependency direction, migration strategy) are only enumerated in the body, not the description; it is clearly above anchor 3's 1-2 actions.

Completeness

Explicitly answers both what ('Choose, review, or migrate Apple-platform architecture and module boundaries') and when with concrete trigger phrases ('Use when feature complexity may justify MVVM, MVI, TCA, Clean Architecture, a Coordinator, or an incremental migration beyond straightforward SwiftUI MV'). Matches anchor 5; anchor 4's 'when could be more explicit' does not apply.

Trigger Term Quality

Strong natural pattern terms ('MVVM, MVI, TCA, Clean Architecture, a Coordinator ... SwiftUI MV, architecture, module boundaries'), but a few phrases users would naturally say are missing — e.g. 'iOS app architecture', 'VIPER', 'app structure'. Fits anchor 4 (good coverage, a few natural terms missing), not anchor 5.

Distinctiveness Conflict Risk

A clear niche — Apple-platform architecture pattern selection — with distinctive pattern-name triggers that are unlikely to fire for sibling skills (SwiftUI wiring, navigation APIs, concurrency, test syntax), which the body explicitly routes elsewhere. Matches anchor 5's clear niche with minimal conflict risk.