スキル fixing-metadata
📦
fixing-metadata
安全
Review and Fix Web Page Metadata
Ensure your web pages have correct, complete metadata for SEO and social sharing. Identifies violations, duplicates, and missing tags with specific fixes to improve search visibility and social card previews.
対応: Claude Codex Code(CC)
1
スキルZIPをダウンロード
2
Claudeでアップロード
設定 → 機能 → スキル → スキルをアップロードへ移動
3
オンにして利用開始
テストする
「fixing-metadata」を使用しています。 Review metadata in page.tsx
期待される結果:
- Violation: Missing meta description on line 15
- Why it matters: Search engines use descriptions for search result snippets
- Fix: Add export const metadata = { description: 'Your page description here' }
- Violation: Duplicate canonical tag at line 23
- Why it matters: Multiple canonicals confuse search engines about preferred URL
- Fix: Remove the manual <link rel="canonical"> and use the metadata API instead
「fixing-metadata」を使用しています。 Check Open Graph setup for blog post page
期待される結果:
- Found: og:title, og:description, and og:image are configured correctly
- Issue: og:url does not match the canonical URL
- Recommendation: Update og:url to match canonical for consistent social sharing
- Note: Twitter card type is summary_large_image which is appropriate for blog posts
セキュリティ監査
安全v1 • 2/17/2026
Static analysis flagged potential external commands and cryptographic patterns, but manual review confirms all findings are false positives. The skill contains only markdown documentation with no executable code. Backticks are used for markdown formatting of command examples, not shell execution. All flagged patterns are SEO terminology in context.
1
スキャンされたファイル
116
解析された行数
3
検出結果
1
総監査数
中リスクの問題 (1)
Markdown backtick usage flagged as shell execution
Static scanner detected backtick characters at lines 12 and 15 as potential Ruby/shell execution. Manual review confirms these are markdown code formatting for skill command syntax (e.g., `/fixing-metadata`), not executable code. This is documentation text, not a security risk.
低リスクの問題 (2)
SKILL.md:3SKILL.md:26SKILL.md:39SKILL.md:52SKILL.md:55SKILL.md:57SKILL.md:62SKILL.md:63SKILL.md:75SKILL.md:113
SEO terminology flagged as weak cryptography
Static scanner flagged multiple instances of 'weak cryptographic algorithm' patterns. Review confirms these are false positives triggered by common SEO words like 'canonical', 'description', and 'page' in markdown text. No cryptographic operations exist in this documentation file.
Web terminology flagged as system reconnaissance
Static scanner detected 'system reconnaissance' patterns in words like 'robots', 'canonical', and 'page'. Manual review confirms these are standard SEO and web development terms used in metadata documentation context. No actual system probing or reconnaissance code exists.
監査者: claude
品質スコア
38
アーキテクチャ
95
保守性
87
コンテンツ
41
コミュニティ
96
セキュリティ
83
仕様準拠
作れるもの
SEO Audit for New Marketing Pages
Review newly created marketing pages for complete metadata including titles, descriptions, canonical URLs, and social sharing tags before launch
Fix Duplicate Metadata Issues
Identify and resolve competing metadata systems or duplicate tags that confuse search engines and social platforms
Implement Structured Data Correctly
Validate JSON-LD structured data blocks to ensure they accurately represent page content without invented ratings or reviews
これらのプロンプトを試す
Quick Metadata Review
/fixing-metadata src/app/layout.tsx
Apply Metadata Guidelines
/fixing-metadata
Audit Social Card Configuration
Review the Open Graph and Twitter card metadata in this component and suggest improvements for better social sharing previews
Fix Canonical and Indexing Issues
Check our metadata configuration for duplicate canonical tags, incorrect robots directives, and noindex pages that should be indexed
ベストプラクティス
- Define metadata in one place per page using your framework's metadata API rather than mixing systems
- Verify social card previews on actual URLs using platform debugging tools before going live
- Keep metadata minimal and focused on actual page content rather than optimizing for every possible tag
回避
- Adding duplicate title or description tags through multiple systems like react-helmet and Next.js metadata API simultaneously
- Using dynamic or random values in metadata that change on each page load or build
- Inventing structured data for ratings, reviews, or prices that do not exist on the actual page
よくある質問
What types of metadata does this skill review?
The skill reviews page titles, meta descriptions, canonical URLs, robots directives, Open Graph tags, Twitter cards, favicons, manifest files, structured data (JSON-LD), and locale/alternate language configuration.
Will this skill automatically fix my metadata issues?
No, the skill identifies violations and provides concrete fix suggestions with code examples, but you must apply the changes yourself. This ensures you understand and approve all metadata modifications.
Can this skill migrate me to a different metadata framework?
No, the skill follows your project's existing metadata pattern whether that's Next.js metadata API, react-helmet, or manual head tags. It does not introduce new frameworks or libraries unless explicitly requested.
How does the skill handle duplicate metadata?
The skill identifies when multiple systems emit the same metadata tags (like duplicate titles or canonicals) and recommends consolidating to a single source to prevent confusing search engines.
What should I do about structured data?
Only add JSON-LD structured data when it clearly maps to real page content. Do not invent ratings, reviews, or organization details. The skill validates that structured data reflects what is actually rendered.
Can this skill test my social card previews?
The skill reviews your metadata configuration for correctness, but to see actual previews you should use platform debugging tools like Facebook Sharing Debugger, Twitter Card Validator, or LinkedIn Post Inspector on live URLs.