hexjasur/AGENTS.md / skills/web-seo
web-seo
UnverifiedImprove discoverability of public web pages with crawlable architecture, unique titles and descriptions, canonical URLs, internal links, structured data, sitemaps, robots rules, and social previews. Use for public routes, SEO audits, indexing or "not showing in Google" problems, migrations and redirects, JavaScript-rendered sites, and share-card previews.
Web SEO
Build for people first, then make the same useful structure clear to crawlers. Fix indexing and rendering problems before polishing metadata, and never promise rankings.
Procedure
- [ ] 1. Inventory public routes and what should (not) be indexed
- [ ] 2. Confirm crawl and render: status, robots, canonical, rendered HTML
- [ ] 3. Titles, descriptions, headings, and content per page
- [ ] 4. Internal links, sitemap, redirects, and international targeting
- [ ] 5. Structured data and social previews that match visible content
- [ ] 6. Validate with real tools and record what was not tested
- List indexable routes, utility pages to exclude, rendering mode (static, SSR, CSR), locales, and the canonical host.
- Every indexable URL returns
200, is not blocked byrobots.txt, has no straynoindex, and exposes its main content and links in rendered HTML. Test with URL Inspection or the rendered DOM, not view-source alone. - Give each page a unique, accurate
<title>(clear and concise), a useful meta description, one primaryh1, and original content that answers the search intent. - Link pages with crawlable
<a href>and descriptive anchor text. Keep an XML sitemap of canonical, indexable URLs with accuratelastmod. Use301or308for permanent moves and avoid redirect chains. - Add JSON-LD only for content visible on the page, using a type Google supports, and validate it. Set Open Graph and Twitter Card tags for share previews.
- Run Lighthouse, Search Console (coverage, enhancements, Core Web Vitals) when available, and a link and redirect crawl.
Guardrails
- Do not keyword-stuff, hide text, cloak, build doorway pages, or mark up content users cannot see.
- Do not block CSS or JavaScript that a page needs to render.
robots.txtcontrols crawling, not indexing or access. To keep a page out of search, allow crawling and usenoindex(meta orX-Robots-Tag), or require authentication.- Do not rely on
rel="canonical",noindex, orhreflangthat contradict each other or the sitemap. - Do not publish scaled, low-value, or unreviewed AI-generated pages to manipulate rankings.
- Do not add structured data or claims you cannot support.
Gotchas
- Content that appears only after a click, scroll, or login may never be indexed. Links built with
onclickor#fragments are not crawled. - Soft 404s (a
200page that says "not found") waste crawl and hurt quality; return404or410. - Parameterized, faceted, and session URLs create duplicates. Canonicalize, restrict, or avoid generating them.
- Changing URLs without redirects discards accumulated signals. Map every old URL to the closest new one and test.
hreflangneeds return links and self-references, and each locale needs its own canonical.- Lazy-loaded images and content need real
<img>andsrc/srcset, or an observer-triggered load that crawlers can trigger. <meta name="keywords">is ignored. Title and description are a snippet suggestion, not a guarantee.- Google rarely shows FAQ and HowTo rich results now; check current feature documentation before investing in a markup type.
- To control use by AI features and crawlers, use the documented robots tokens and
nosnippetordata-nosnippetrules. There is no special markup required to appear in AI features beyond being indexable and useful.
Report format
For each issue: URL or pattern, evidence, impact, and fix, ordered by how many pages it affects. State what was inspected (rendered HTML, Search Console, crawl) and what was not.
Read references/seo.md for the per-page checklist, status-code table, structured data guidance, and sources.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 193k | Updated | Claude CodeCodex | Verified | Install |