` in `<head>` - Missing meta description → add `<meta name="description">` - Missing OG tags → generate from page content - Missing H1 → add heading from page content - Missing alt text → add descriptive alt attributes - Missing canonical → add `<link rel="canonical">` - Missing sitemap → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/sitemap-analyzer.mjs generate <dir> <url>` - Missing robots.txt → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/robots-analyzer.mjs generate <dir> <url>` **GEO fixes**: - Missing llms.txt → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/llms-txt-generator.mjs <dir>` - Missing structured data → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/schema-engine.mjs generate <type> --out=<path>` - Blocked AI bots → update robots.txt to allow GPTBot, PerplexityBot, ClaudeBot, Google-Extended - Improve content structure: add question-based H2s, TL;DR summaries, definition patterns **AEO fixes**: - Missing FAQ schema → generate FAQPage from Q&A content on page - Missing speakable → add speakable schema for key content sections - Snippet optimization → restructure answers to 40-60 word direct answers **Performance fixes**: - Missing image dimensions → add `width` and `height` attributes to all `<img>` tags - Render-blocking scripts → add `async` or `defer` to third-party `<script>` tags - Font display issues → add `font-display: swap` to `@font-face` declarations - Lazy loading → add `loading="lazy"` to below-fold images - LCP priority → add `fetchpriority="high"` to the LCP image (hero/above-fold) - Missing preconnect → add `<link rel="preconnect">` hints for third-party origins (fonts, CDNs, analytics) **Security fixes**: - Missing CSP → add `<meta http-equiv="Content-Security-Policy" content="...">` in `<head>` - Missing referrer policy → add `<meta name="referrer" content="strict-origin-when-cross-origin">` - Unsafe target="_blank" → add `rel="noopener noreferrer"` to all `target="_blank"` links - Unsandboxed iframes → add `sandbox` attribute to `<iframe>` elements - Mixed content → fix `http://` URLs to `https://` in src, href, and content attributes **Content fixes**: - Long paragraphs → break up paragraphs longer than 150 words into smaller chunks - Complex sentences → simplify sentences with Flesch-Kincaid grade level > 12 - Passive voice → rewrite passive voice constructions to active voice **Vertical fixes**: - Missing Product schema → add Product JSON-LD for e-commerce product pages - Missing LocalBusiness schema → add LocalBusiness JSON-LD with business details - Missing NAP data → add Name, Address, Phone in structured data and visible on page - Missing opening hours → add `openingHoursSpecification` to LocalBusiness schema ## Phase 4: Verify Re-run all eight scanners. Show before/after score comparison for each category. ## Phase 5: Save History ```bash node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> seo <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> geo <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> aeo <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> citability <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> content <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> performance <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> vertical <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> security <score> ``` ## Phase 6: Content Strategy Advise on content optimizations the scanner cannot automate: - Rewrite H2 headers as questions for AI citation - Add concise TL;DR summaries under key sections - Write in citation-ready format: factual, quotable, 134-167 word passages - Create comparison tables, statistics pages, glossaries ## Phase 7: Backlink Strategy Guide link building: create link-worthy assets, guest posting, broken link building, digital PR, expert roundups. ## Phase 8: Search Console Action Plan After fixing issues, guide the user through submitting their improved site to search engines. ### Google Search Console (GSC) 1. **Submit Sitemap** — GSC → Sitemaps → Enter `sitemap.xml` → Submit 2. **Request Indexing for Money Pages** — URL Inspection → paste each fixed page → Request Indexing (10-12/day limit, prioritize homepage, pricing, top landing pages) 3. **Check Index Coverage** — Pages → review "Not indexed" list. "Crawled - currently not indexed" = needs content/links. "Discovered - currently not indexed" = needs internal links. 4. **Validate Robots.txt** — Settings → Crawling → verify AI bots are unblocked 5. **Check Rich Results** — Enhancements → review schema types. Test with [Rich Results Test](https://search.google.com/test/rich-results). 6. **Monitor Core Web Vitals** — Experience → Core Web Vitals → fix "Poor" URLs (LCP, CLS, INP) ### Bing Webmaster Tools 1. **Submit Sitemap** — Configure Sitemaps → submit sitemap.xml URL 2. **Submit URLs** — URL Submission → submit top pages (10,000/day limit) 3. **Enable IndexNow** — Generate key at indexnow.org → place key file at domain root. Feeds Bing, Yandex, Seznam. 4. **Verify Robots.txt** — Configure My Site → Block URLs → Robots.txt Tester. Bingbot feeds Copilot and ChatGPT Browse. ### AI Search Verification 1. **Test AI Visibility** (2-4 weeks after deploy) — search brand + keywords in ChatGPT, Perplexity, Gemini, Google AI Overviews 2. **Verify llms.txt** — confirm `https://yourdomain.com/llms.txt` returns content 3. **Monitor AI Citations Weekly** — track which pages get cited vs competitors in Perplexity and ChatGPT ## Phase 9: Next Steps Recommend which `/claude-rank` sub-commands to run next based on lowest scores. Prioritize by score: - If **SEO < 60**, run `/claude-rank:rank-seo` for detailed on-page analysis - If **GEO < 60**, run `/claude-rank:rank-geo` for AI visibility deep-dive - If **AEO < 60**, run `/claude-rank:rank-aeo` for snippet and voice optimization - If **Citability < 50**, run `/claude-rank:rank-citability` for detailed per-page citation breakdown across all 7 dimensions - If **Content < 50**, run `/claude-rank:rank-content` for readability analysis and thin page identification - If **Performance < 50**, run `/claude-rank:rank-performance` for CLS/LCP/blocking resource analysis - If **Vertical < 50**, run `/claude-rank:rank-vertical` for industry-specific schema and optimization - If **Security < 50**, run `/claude-rank:rank-security` for header analysis and CSP configuration Present the user with a prioritized action checklist ordered by impact (lowest scores first).

--- name: rank-audit description: Full SEO/GEO/AEO/Citability/Content/Performance/Vertical/Security audit with auto-fix. Scans, reports, fixes, and verifies. --- # Full Audit Comprehensive search optimization audit across 8 dimensions. Finds issues AND fixes them. ## Phase 1: Parallel Scan Run all eight scanners: ```bash node ${CLAUDE_PLUGIN_ROOT}/tools/seo-scanner.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/geo-scanner.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/aeo-scanner.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/citability-scorer.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/content-analyzer.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/perf-scanner.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/vertical-scanner.mjs node ${CLAUDE_PLUGIN_ROOT}/tools/security-scanner.mjs ``` Parse JSON output from each for findings and scores. ## Phase 2: Report Present findings grouped by all 8 categories with severity. Show scores table: | Category | Score | |-------------|-------| | SEO | -- | | GEO | -- | | AEO | -- | | Citability | -- | | Content | -- | | Performance | -- | | Vertical | -- | | Security | -- | | **Overall** | -- | Show summary: X critical, Y high, Z medium, W low across all categories. ## Phase 3: Auto-Fix For each finding, apply the appropriate fix: **SEO fixes** (Edit tool on HTML files): - Missing title → add `` in `<head>` - Missing meta description → add `<meta name="description">` - Missing OG tags → generate from page content - Missing H1 → add heading from page content - Missing alt text → add descriptive alt attributes - Missing canonical → add `<link rel="canonical">` - Missing sitemap → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/sitemap-analyzer.mjs generate <dir> <url>` - Missing robots.txt → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/robots-analyzer.mjs generate <dir> <url>` **GEO fixes**: - Missing llms.txt → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/llms-txt-generator.mjs <dir>` - Missing structured data → run: `node ${CLAUDE_PLUGIN_ROOT}/tools/schema-engine.mjs generate <type> --out=<path>` - Blocked AI bots → update robots.txt to allow GPTBot, PerplexityBot, ClaudeBot, Google-Extended - Improve content structure: add question-based H2s, TL;DR summaries, definition patterns **AEO fixes**: - Missing FAQ schema → generate FAQPage from Q&A content on page - Missing speakable → add speakable schema for key content sections - Snippet optimization → restructure answers to 40-60 word direct answers **Performance fixes**: - Missing image dimensions → add `width` and `height` attributes to all `<img>` tags - Render-blocking scripts → add `async` or `defer` to third-party `<script>` tags - Font display issues → add `font-display: swap` to `@font-face` declarations - Lazy loading → add `loading="lazy"` to below-fold images - LCP priority → add `fetchpriority="high"` to the LCP image (hero/above-fold) - Missing preconnect → add `<link rel="preconnect">` hints for third-party origins (fonts, CDNs, analytics) **Security fixes**: - Missing CSP → add `<meta http-equiv="Content-Security-Policy" content="...">` in `<head>` - Missing referrer policy → add `<meta name="referrer" content="strict-origin-when-cross-origin">` - Unsafe target="_blank" → add `rel="noopener noreferrer"` to all `target="_blank"` links - Unsandboxed iframes → add `sandbox` attribute to `<iframe>` elements - Mixed content → fix `http://` URLs to `https://` in src, href, and content attributes **Content fixes**: - Long paragraphs → break up paragraphs longer than 150 words into smaller chunks - Complex sentences → simplify sentences with Flesch-Kincaid grade level > 12 - Passive voice → rewrite passive voice constructions to active voice **Vertical fixes**: - Missing Product schema → add Product JSON-LD for e-commerce product pages - Missing LocalBusiness schema → add LocalBusiness JSON-LD with business details - Missing NAP data → add Name, Address, Phone in structured data and visible on page - Missing opening hours → add `openingHoursSpecification` to LocalBusiness schema ## Phase 4: Verify Re-run all eight scanners. Show before/after score comparison for each category. ## Phase 5: Save History ```bash node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> seo <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> geo <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> aeo <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> citability <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> content <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> performance <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> vertical <score> node ${CLAUDE_PLUGIN_ROOT}/tools/audit-history.mjs save <dir> security <score> ``` ## Phase 6: Content Strategy Advise on content optimizations the scanner cannot automate: - Rewrite H2 headers as questions for AI citation - Add concise TL;DR summaries under key sections - Write in citation-ready format: factual, quotable, 134-167 word passages - Create comparison tables, statistics pages, glossaries ## Phase 7: Backlink Strategy Guide link building: create link-worthy assets, guest posting, broken link building, digital PR, expert roundups. ## Phase 8: Search Console Action Plan After fixing issues, guide the user through submitting their improved site to search engines. ### Google Search Console (GSC) 1. **Submit Sitemap** — GSC → Sitemaps → Enter `sitemap.xml` → Submit 2. **Request Indexing for Money Pages** — URL Inspection → paste each fixed page → Request Indexing (10-12/day limit, prioritize homepage, pricing, top landing pages) 3. **Check Index Coverage** — Pages → review "Not indexed" list. "Crawled - currently not indexed" = needs content/links. "Discovered - currently not indexed" = needs internal links. 4. **Validate Robots.txt** — Settings → Crawling → verify AI bots are unblocked 5. **Check Rich Results** — Enhancements → review schema types. Test with [Rich Results Test](https://search.google.com/test/rich-results). 6. **Monitor Core Web Vitals** — Experience → Core Web Vitals → fix "Poor" URLs (LCP, CLS, INP) ### Bing Webmaster Tools 1. **Submit Sitemap** — Configure Sitemaps → submit sitemap.xml URL 2. **Submit URLs** — URL Submission → submit top pages (10,000/day limit) 3. **Enable IndexNow** — Generate key at indexnow.org → place key file at domain root. Feeds Bing, Yandex, Seznam. 4. **Verify Robots.txt** — Configure My Site → Block URLs → Robots.txt Tester. Bingbot feeds Copilot and ChatGPT Browse. ### AI Search Verification 1. **Test AI Visibility** (2-4 weeks after deploy) — search brand + keywords in ChatGPT, Perplexity, Gemini, Google AI Overviews 2. **Verify llms.txt** — confirm `https://yourdomain.com/llms.txt` returns content 3. **Monitor AI Citations Weekly** — track which pages get cited vs competitors in Perplexity and ChatGPT ## Phase 9: Next Steps Recommend which `/claude-rank` sub-commands to run next based on lowest scores. Prioritize by score: - If **SEO < 60**, run `/claude-rank:rank-seo` for detailed on-page analysis - If **GEO < 60**, run `/claude-rank:rank-geo` for AI visibility deep-dive - If **AEO < 60**, run `/claude-rank:rank-aeo` for snippet and voice optimization - If **Citability < 50**, run `/claude-rank:rank-citability` for detailed per-page citation breakdown across all 7 dimensions - If **Content < 50**, run `/claude-rank:rank-content` for readability analysis and thin page identification - If **Performance < 50**, run `/claude-rank:rank-performance` for CLS/LCP/blocking resource analysis - If **Vertical < 50**, run `/claude-rank:rank-vertical` for industry-specific schema and optimization - If **Security < 50**, run `/claude-rank:rank-security` for header analysis and CSP configuration Present the user with a prioritized action checklist ordered by impact (lowest scores first).