SwiftUI Expert Skill
Actionable SwiftUI guidance for state management, performance, modern APIs, and AI-assisted code review.
What does this skill do, and when should you use it?
SwiftUI Expert Skill follows the open Agent Skills format and supplies concise, on-demand references for SwiftUI development and review. It covers state management, view composition, lists, navigation, animation, Swift Charts, macOS, accessibility, image optimization, and iOS 26+ Liquid Glass. The repository also includes a Python and xctrace toolchain for recording and analyzing Instruments traces. It is aimed at teams and developers working with modern SwiftUI code rather than at prescribing an architecture or code style.
Provides a SKILL.md workflow and topic-specific Markdown references; guides agents through SwiftUI state, view structure, performance, lists, navigation, animation, charts, macOS, accessibility, and related APIs; wraps xctrace with scripts/record_trace.py for trace recording; uses scripts/analyze_trace.py to analyze Time Profiler, Hangs, Animation Hitches, SwiftUI updates, and cause graphs; emits JSON and Markdown analysis; includes an agents/openai.yaml manifest and installation examples for several agent tools.
- A team adopting modern SwiftUI APIs wants consistent defaults for code review.
- A developer is refactoring SwiftUI views or data flow and wants state-management and composition guidance.
- An app has performance issues in lists, scrolling, sheets, or navigation and needs targeted profiling guidance.
- A developer has an existing `.trace` file and wants structured analysis of hangs, hitches, or costly SwiftUI updates.
- A project needs guidance for Swift Charts, accessibility, macOS scenes, or iOS 26+ Liquid Glass.
What are this skill's strengths and limitations?
- Covers many practical SwiftUI development, review, and performance-diagnosis topics.
- Loads reference material on demand to avoid bloating task context.
- Includes a Python toolchain for recording and analyzing Instruments traces.
- Uses the MIT License and documents integrations with multiple agent tools.
- The supplied source does not include SKILL.md, so its frontmatter and complete workflow cannot be independently verified here.
- Trace recording and analysis require xctrace and Xcode on macOS.
- The maintenance workflow requires the Sosumi MCP to be available.
- The repository metadata lists no topics; the stated scope is based primarily on the README.
How do you install this skill?
With skills.sh, run npx skills add https://github.com/avdlee/swiftui-agent-skill --skill swiftui-expert-skill. For Codex/OpenAI-compatible tools, copy or symlink swiftui-expert-skill/ into $CODEX_HOME/skills/swiftui-expert-skill. The README also documents Claude Code, Cursor, pi, and manual installation; follow the target tool’s official instructions for the final skill location.
How do you use this skill?
In a compatible agent, ask it to use swiftui-expert, for example: Use the swiftui expert skill and review the current SwiftUI code for state-management and performance improvements. When a .trace path is supplied, ask the agent to analyze it; for a new profiling session, ask it to record a trace.