jpskill.com
💬 コミュニケーション コミュニティ

mobile

Root mobile: native vs cross-platform decisions, app store guidelines, mobile UX principles

⚡ おすすめ: コマンド1行でインストール(60秒)

下記のコマンドをコピーしてターミナル(Mac/Linux)または PowerShell(Windows)に貼り付けてください。 ダウンロード → 解凍 → 配置まで全自動。

🍎 Mac / 🐧 Linux
mkdir -p ~/.claude/skills && cd ~/.claude/skills && curl -L -o mobile.zip https://jpskill.com/download/22264.zip && unzip -o mobile.zip && rm mobile.zip
🪟 Windows (PowerShell)
$d = "$env:USERPROFILE\.claude\skills"; ni -Force -ItemType Directory $d | Out-Null; iwr https://jpskill.com/download/22264.zip -OutFile "$d\mobile.zip"; Expand-Archive "$d\mobile.zip" -DestinationPath $d -Force; ri "$d\mobile.zip"

完了後、Claude Code を再起動 → 普通に「動画プロンプト作って」のように話しかけるだけで自動発動します。

💾 手動でダウンロードしたい(コマンドが難しい人向け)
  1. 1. 下の青いボタンを押して mobile.zip をダウンロード
  2. 2. ZIPファイルをダブルクリックで解凍 → mobile フォルダができる
  3. 3. そのフォルダを C:\Users\あなたの名前\.claude\skills\(Win)または ~/.claude/skills/(Mac)へ移動
  4. 4. Claude Code を再起動

⚠️ ダウンロード・利用は自己責任でお願いします。当サイトは内容・動作・安全性について責任を負いません。

🎯 このSkillでできること

下記の説明文を読むと、このSkillがあなたに何をしてくれるかが分かります。Claudeにこの分野の依頼をすると、自動で発動します。

📦 インストール方法 (3ステップ)

  1. 1. 上の「ダウンロード」ボタンを押して .skill ファイルを取得
  2. 2. ファイル名の拡張子を .skill から .zip に変えて展開(macは自動展開可)
  3. 3. 展開してできたフォルダを、ホームフォルダの .claude/skills/ に置く
    • · macOS / Linux: ~/.claude/skills/
    • · Windows: %USERPROFILE%\.claude\skills\

Claude Code を再起動すれば完了。「このSkillを使って…」と話しかけなくても、関連する依頼で自動的に呼び出されます。

詳しい使い方ガイドを見る →
最終更新
2026-05-18
取得日時
2026-05-18
同梱ファイル
1
📖 Claude が読む原文 SKILL.md(中身を展開)

この本文は AI(Claude)が読むための原文(英語または中国語)です。日本語訳は順次追加中。

mobile

Purpose

This skill helps OpenClaw assist in mobile app development by evaluating native (e.g., Swift for iOS, Kotlin for Android) vs. cross-platform (e.g., React Native, Flutter) approaches, ensuring compliance with app store guidelines (e.g., Apple's App Store Review Guidelines, Google Play Policies), and applying mobile UX principles like touch-friendly designs and responsive layouts.

When to Use

Use this skill when starting a new mobile project, refactoring existing apps, or optimizing for platforms; for example, when deciding between native for performance-critical apps or cross-platform for faster development. Apply it during design phases for UX audits or when preparing app submissions to avoid rejections.

Key Capabilities

  • Analyze native vs. cross-platform trade-offs: Compare build times, maintenance, and performance using metrics like app size and startup speed.
  • Validate against app store guidelines: Check for issues like data privacy compliance (e.g., GDPR alignment) or in-app purchase rules via automated scans.
  • Enforce mobile UX principles: Generate recommendations for layouts, such as ensuring button sizes > 44x44 pixels for iOS, or using Material Design components for Android.
  • Provide code generation: Output snippets for platform-specific features, like handling permissions in Android Manifest or SwiftUI views.
  • Integration with tools: Query external APIs for real-time guideline updates, e.g., fetching Apple guidelines via unofficial scrapers (not recommended for production).

Usage Patterns

Invoke this skill via OpenClaw's CLI or API by prefixing commands with the skill ID, e.g., openclaw mobile [subcommand]. Always pass required parameters like platform and project type. For decisions, use interactive mode to refine inputs based on user feedback. Example: Pipe outputs to other skills for chaining, like combining with a "testing" skill for automated UX tests. Handle asynchronous operations by checking response status codes.

Common Commands/API

Use the OpenClaw CLI for quick tasks or the REST API for programmatic access. Authentication requires setting $OPENCLAW_API_KEY as an environment variable.

  • CLI Command: openclaw mobile decide --platform ios --type native-vs-cross --factors performance,cost

    • Output: JSON with pros/cons, e.g., {"native": "Faster performance but higher cost", "cross": "Shared codebase but potential UI inconsistencies"}.
    • Code Snippet:
      response = subprocess.run(['openclaw', 'mobile', 'decide', '--platform', 'android', '--type', 'cross'], capture_output=True)
      print(response.stdout.decode())
  • CLI Command: openclaw mobile validate --app-path /path/to/app --store apple

    • Flags: --store (apple/google) for guideline checks; --app-path for local file analysis.
    • Code Snippet:
      import os
      os.system('openclaw mobile validate --app-path ./myapp --store google')
      # Checks for policies like ad compliance and returns exit code 0 if pass
  • API Endpoint: POST /api/v1/skills/mobile/decide

    • Body: JSON like {"platform": "ios", "type": "ux-principles", "query": "button sizing"}
    • Response: 200 OK with data, e.g., {"recommendation": "Use .frame(width: 50, height: 50) for buttons"}.
    • Code Snippet:
      import requests
      headers = {'Authorization': f'Bearer {os.environ.get("OPENCLAW_API_KEY")}'}
      response = requests.post('https://api.openclaw.ai/api/v1/skills/mobile/decide', json={"platform": "android"}, headers=headers)
      print(response.json()['advice'])
  • Config Format: Use YAML for custom profiles, e.g.,

    mobile:
      defaultPlatform: android
      uxRules:
        minButtonSize: 48

    Load via: openclaw mobile config --file path/to/config.yaml.

Integration Notes

Integrate by wrapping OpenClaw calls in your build scripts or IDE extensions. For example, in a CI/CD pipeline, add a step like openclaw mobile validate before deployment. If using VS Code, install the OpenClaw extension and bind to keyboard shortcuts. For API integrations, ensure $OPENCLAW_API_KEY is set securely via secrets management. Handle rate limits by caching responses; OpenClaw enforces 100 requests/min, so implement retry logic with exponential backoff.

Error Handling

Common errors include authentication failures (HTTP 401) if $OPENCLAW_API_KEY is missing or invalid—fix by verifying the env var and regenerating keys. For invalid inputs, like unsupported platforms, expect HTTP 400 with details; parse the error JSON and prompt for corrections. Platform-specific issues, e.g., iOS guideline mismatches, return structured errors like {"code": "APP_STORE_101", "message": "Privacy policy missing"}—use try-except blocks in scripts:

  try:
      result = requests.post(url, headers=headers)
      result.raise_for_status()
  except requests.exceptions.HTTPError as e:
      print(f"Error: {e.response.json()['message']}. Retrying...")

Always log errors with timestamps for debugging, and fallback to default behaviors if API is down.

Graph Relationships

  • Related to: "ios" skill (child node for iOS-specific implementations)
  • Related to: "android" skill (sibling node for Android optimizations)
  • Connected via: "ux-design" cluster (parent for shared UX principles)
  • Links to: "app-store" external resource for guideline references