swiftui-expert-skill
Write, review, or improve SwiftUI code following best practices for state management, view composition, performance, macOS-specific APIs, and iOS 26+ Liquid Glass adoption. Use when building new SwiftUI features, refactoring existing views, reviewing code quality, or adopting modern SwiftUI patterns. Also triggers whenever an Xcode Instruments `.trace` file is referenced (to analyse it) or the user asks to **record** a new trace — attach to a running app, launch one fresh, or capture a manually-stopped session with the bundled `record_trace.py`. A target SwiftUI source file is optional; if provided it grounds recommendations in specific lines, but a trace alone is enough to diagnose hangs, hitches, CPU hotspots, and high-severity SwiftUI updates.
95
92%
Does it follow best practices?
Impact
100%
1.16xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly defines a specific niche (SwiftUI development + Xcode Instruments performance analysis), lists concrete actions, and provides explicit trigger conditions. It covers both the code quality/writing aspect and the performance diagnostics aspect with natural keywords users would employ. The description is detailed without being padded, and uses appropriate third-person voice throughout.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: write/review/improve SwiftUI code, analyze .trace files, record new traces, attach to running apps, launch fresh, capture manually-stopped sessions, diagnose hangs/hitches/CPU hotspots/high-severity SwiftUI updates. Very detailed and actionable. | 3 / 3 |
Completeness | Clearly answers both 'what' (write/review/improve SwiftUI code, analyze trace files, record traces, diagnose performance issues) and 'when' ('Use when building new SwiftUI features, refactoring existing views, reviewing code quality, or adopting modern SwiftUI patterns. Also triggers whenever an Xcode Instruments .trace file is referenced...'). Explicit trigger guidance is provided. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms: 'SwiftUI', 'Xcode Instruments', '.trace file', 'macOS', 'iOS 26+', 'Liquid Glass', 'state management', 'view composition', 'refactoring', 'hangs', 'hitches', 'CPU hotspots', 'record_trace.py'. These are terms users would naturally use when requesting help in this domain. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche combining SwiftUI development with Xcode Instruments trace analysis. The specific mention of iOS 26+ Liquid Glass, macOS-specific APIs, record_trace.py, and .trace files makes this very unlikely to conflict with generic coding or other platform-specific skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, comprehensive skill that effectively balances overview-level guidance with deep reference material through progressive disclosure. The trace recording and analysis workflows are particularly strong with concrete commands, decision logic, and interpretation criteria. The main weakness is moderate redundancy between the Topic Router table and the References section, plus a few unnecessary operating rules that state what Claude would do by default.
Suggestions
Consolidate the Topic Router table and the References section into a single table to eliminate redundancy — the References section adds only minor extra description that could be folded into the Topic Router.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but has notable redundancy: the Topic Router table and the References section list the same files with nearly identical descriptions. The trace recording and analysis workflows are detailed but appropriately so given their complexity. Some sections like the Operating Rules contain guidance Claude would naturally follow (e.g., 'Follow Apple's Human Interface Guidelines'). | 2 / 3 |
Actionability | The skill provides fully executable bash commands for trace recording and analysis, a concrete correctness checklist with specific rules, clear workflow steps with exact script invocations and flags, and specific decision logic (e.g., simulator → Time Profiler, real device → SwiftUI template). The Topic Router directs to specific reference files for each concern. | 3 / 3 |
Workflow Clarity | Multi-step workflows are clearly sequenced with explicit validation and decision points. The trace recording workflow has numbered steps with conditional branching (agent-driven vs interactive), stop-file signaling, and flows into the analysis workflow. The trace analysis workflow includes scoping decisions, window resolution, interpretation criteria (coverage % thresholds), and a prioritized output plan. The correctness checklist serves as a validation checkpoint for code tasks. | 3 / 3 |
Progressive Disclosure | The skill is an excellent overview that delegates detailed content to 20+ well-organized reference files, all one level deep. The Topic Router table provides clear navigation. Inline content is limited to workflow steps and the correctness checklist — exactly what belongs in the top-level file. References are clearly signaled with consistent formatting. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- avdlee/swiftui-agent-skill
- Commit
a5f130a
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.