What problem does it solve? AI-generated SwiftUI text code frequently contains deprecated patterns like Text + Text concatenation, hardcoded font sizes that break Dynamic Type, and invented APIs such as .fontSize() that never existed. This Skill audits a macOS SwiftUI codebase for nine specific typography defects and reports or fixes them with evidence grounded in real shipping apps. ## Core Features & Use Cases - Nine-rule defect detection: Covers deprecated Text + Text concatenation, the design-only Font.system overload, hardcoded font sizes, missing lineLimit space reservation, malformed AttributedString, mis-gated TextRenderer, hand-rolled label rows, jiggling numerics without monospacedDigit(), and hallucinated type APIs. - Hybrid lint engine: Combines tier-1 grep tells with a tier-2 ast-grep structural rule for multi-line Text + Text, emitting unified JSON and SARIF output with parse-probe warnings. - Evidence-grounded fixes: Verifies uncertain findings against a corpus of 1,857 shipping macOS apps via the swiftui-ctx CLI and Apple documentation via Sosumi, then applies safe auto-fixes (e.g., appending weight: .regular) under a fix-safety protocol. - Use Case: Point the Skill at an in-progress macOS SwiftUI project to receive structured findings in swiftui-audits/typography-text/, each citing a real GitHub permalink and Apple doc, with mechanical defects fixed automatically. ## Quick Start Audit the SwiftUI sources in my macOS project for typography and text defects and write the findings to the swiftui-audits folder.
