What does this skill do, and when should you use it?
SwiftUI Pro is an Agent Skill for Claude Code, Codex, and other AI coding tools. It reviews Swift and SwiftUI projects against guidance for modern APIs, view structure, data flow, navigation, design, accessibility, performance, Swift, and code hygiene. Findings are organized by file and include locations, violated rules, and short before/after fixes. It fits developers who want AI-assisted SwiftUI review, but the source does not provide a test suite or independent evidence of runtime validation.
It reads Swift and SwiftUI code and checks it against the skill's api, views, data, navigation, design, accessibility, performance, swift, and hygiene references. It looks for deprecated APIs, data-flow problems, navigation and performance issues, and accessibility concerns involving Dynamic Type, VoiceOver, and Reduce Motion. It produces file-based findings with relevant lines, rule names, concise before/after fixes, and a prioritized summary. The supplied source does not show it compiling code, running tests, or invoking external commands.
- A SwiftUI developer wants a pre-commit review of modern API usage and project conventions.
- A team maintaining an older SwiftUI codebase needs to identify deprecated APIs and modernize them.
- A developer addressing VoiceOver, Dynamic Type, or Reduce Motion needs a focused accessibility review.
- A developer investigating slow updates or complicated state management needs checks for view structure, data flow, and performance.
- A developer using an AI coding assistant to generate SwiftUI wants review focused on common AI-generated mistakes.
How do you install this skill?
- This is a static documentation review; the skill was not executed, and its accuracy or completeness on real SwiftUI projects was not verified.
- The guidance assumes iOS 26 and Swift 6.2; applicability should be checked for older deployment targets or cross-platform SwiftUI projects.
- Failure handling and boundaries for sensitive-code review are not explicit; manually review advice involving secrets, private data, network calls, or destructive changes.
- No evidence is provided for Chinese-language support or mainland-China network reachability.
- Local filesystem
Node.js for npx-based installation
Install the skill with npx:
npx skills add https://github.com/twostraws/swiftui-agent-skill --skill swiftui-pro
Claude Code can also install it with:
/plugin marketplace add twostraws/SwiftUI-Agent-Skill
/plugin install swiftui-pro@swiftui-agent-skill
The source says installation can be scoped to one project or made available across projects, but does not document a client-specific installation directory.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/twostraws/SwiftUI-Agent-Skill.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/swiftui-pro" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Use the SwiftUI Pro skill to look for performance problems in this project.
- $swiftui-pro Focus on accessibility
- /swiftui-pro Check for deprecated API
Trigger it in Claude Code with /swiftui-pro or in Codex with $swiftui-pro. You can also use natural language, for example: “Use the SwiftUI Pro skill to look for performance problems in this project.” For a partial review, specify the focus, such as “$swiftui-pro Focus on accessibility” or “/swiftui-pro Check for deprecated API”.
What are this skill's strengths and limitations?
- Covers APIs, views, data flow, navigation, design, accessibility, performance, Swift, and maintainability.
- Explicitly instructs the reviewer to report genuine problems rather than nitpick.
- Defines a practical output format with files, lines, rules, and before/after examples.
- Sets clear baselines for iOS 26+, Swift 6.2+, and modern Swift concurrency.
- Available under the MIT License.
- The skill references multiple reference files whose contents are not included in the supplied material.
- The source does not show compilation, test execution, or runtime validation, so it is not a substitute for those checks.
- No test suite, performance benchmark, or independent platform-validation evidence is provided.
- Review quality depends on having enough project context; partial reviews necessarily cover less.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| SwiftUI Pro this page | 57 · Use with care | ★ 5.9k | 19d ago | MIT |
| SwiftUI Pro | 51 · Use with care | ★ 5.9k | 19d ago | MIT |
| Swift Concurrency Pro | 65 · Recommended | ★ 572 | 5mo ago | MIT |
| Swift Concurrency Pro | 60 · Recommended | ★ 572 | 5mo ago | MIT |
| SwiftUI Expert Skill | Not yet reviewed | ★ 3.7k | 21d ago | MIT |
How did FollowSkills review this skill?
The skill provides static code-review guidance only; it runs no commands, requires no external service or elevated permission, explicitly asks before adding third-party frameworks, and warns against committing secrets. It does not explain user-code or sensitive-data flows, review side effects, rollback, or result verification, so 7 points are deducted.
The review steps, reference files, and output format are broadly consistent and cover APIs, data flow, navigation, accessibility, performance, and hygiene. The supplied material lacks abnormal-input handling, failure feedback, key-path tests, and reproducible execution evidence; 12 points are deducted under static calibration.
The intended audience, SwiftUI scenarios, full-review and partial-review modes, and invocation examples are reasonably clear. Non-fit boundaries, input/output limits, Chinese-language support, and mainland-China environment reachability are not established; 4 points are deducted.
The skill uses layered reference files and includes version, license, examples, and some usage guidance; repository context also supplies installation and contribution notes. Maintenance ownership, changelog, versioning policy, dependency versions, and troubleshooting paths are incomplete, so 5 points are deducted.
The rules address modern SwiftUI APIs, accessibility, data flow, navigation, and performance, and the example output is directly actionable. No execution, test suite, or representative review result is supplied, so correctness and completeness still require human review; 9 points are deducted.
The supplied files are auditable and the rules and output requirements are explicit, but there is no committed test suite, CI result, independent corroboration, or source citation for key judgments. Six points are deducted, with the static ceiling limiting this dimension to 5.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →