Seo technical audit
Curated SEO Agent Skills pack for AI agents (Agent Skills open standard): technical audit, content briefs, JSON-LD schema, GEO/AEO, and anti-spam quality gates.
npx -y skills add gaguero/seo-agent-skills --skill seo-technical-auditAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Audits the technical SEO health of a page or site: crawlability, indexing, rendering (JS), HTTP status codes, redirects, robots.txt, XML sitemaps, canonical tags, pagination, soft 404s, hreflang / international targeting, site architecture and internal-link depth, plus Core Web Vitals (LCP, INP, CLS) and mobile-friendliness. Produces a severity-ranked findings list with evidence and exact fixes. Use when someone asks why pages aren't indexed or ranking, mentions crawl errors, robots/sitemap/canonical/redirect/hreflang issues, page speed, Core Web Vitals, or wants a technical SEO audit.
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
5.7 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it
Technical SEO Audit
Technical SEO is the foundation: if a page can't be crawled, rendered, and indexed, nothing else (content, schema, AI visibility) can help it. Audit in that dependency order. Label every finding measured (you observed it) or inferred (deduced without direct access).
What to collect first
- The URL(s) or domain, and the target market/language(s).
- What you can actually inspect: live HTTP response + headers, rendered DOM,
robots.txt, sitemap(s), Lighthouse/PSI report, Search Console coverage. - If you only have pasted HTML, say so and scope findings accordingly.
Audit checklist (in dependency order)
1. Crawlability
robots.txt: reachable (200)? Does it accidentallyDisallowimportant paths? Is theSitemap:line present and correct? Never block CSS/JS needed to render.- Crawl traps: infinite faceted URLs, session IDs, calendar links, parameter
explosions. Recommend parameter handling /
robotsrules, not blanket blocks. - Internal-link reachability: every important page should be reachable within a few clicks from the home/hub. Flag orphan pages and deep (>3–4 click) pages.
2. Indexability
<meta name="robots">andX-Robots-Tagheader: catch accidentalnoindex.- Canonical: each page declares a self-referential canonical unless intentionally
consolidating. Flag conflicting/cross-domain/chained canonicals and canonicals
that point to non-200 or
noindexURLs. - Status codes: 200 for indexable content; 301 for permanent moves; avoid 302 for permanent; fix 4xx/5xx; eliminate redirect chains and loops.
- Soft 404: "not found" content returned with a 200. Flag thin/empty/error pages that should return 404/410.
- Duplicate content: parameterized, http/https, www/non-www, trailing-slash, and uppercase variants resolving to the same content without canonical/redirect.
3. Rendering
- Is critical content/links present in the initial HTML, or injected by JS? Client-side-only content risks delayed or missed indexing. Recommend SSR/SSG or pre-rendering for primary content and links.
- Compare raw HTML vs rendered DOM when possible. Flag content that only appears after interaction.
4. XML sitemaps
- Valid XML, < 50,000 URLs and < 50 MB uncompressed per file (use a sitemap index beyond that). Only canonical, indexable, 200 URLs — no noindex/redirected/404.
- Submitted in Search Console and referenced from
robots.txt.
5. International / hreflang (multilingual sites)
hreflangset is reciprocal (every alternate links back) and self- referential. Use correct ISO codes (en,es,es-PA, plusx-default).- hreflang URLs must be canonical, indexable 200 pages. Don't mix hreflang with conflicting canonicals across languages. This is the #1 multilingual SEO bug.
6. Site architecture
- Logical hierarchy: home → hub/category → detail. Shallow, descriptive URLs.
- Strong internal linking from authority pages to priority pages with descriptive anchor text. Flag important pages with too few internal links.
7. Performance — Core Web Vitals
Field data (CrUX/PSI) is the ground truth; lab (Lighthouse) is diagnostic. A page passes only if all three are good at the 75th percentile:
| Metric | Good | Common causes when failing |
|---|---|---|
| LCP ≤ 2.5 s | loading | slow server/TTFB, render-blocking CSS/JS, unoptimized hero image, no preload |
| INP ≤ 200 ms | interactivity | heavy/long JS tasks, large main-thread work, unoptimized event handlers |
| CLS ≤ 0.1 | stability | images/embeds without dimensions, injected banners, late-loading fonts (FOUT) |
Quote thresholds exactly. Report measured field values when available; otherwise diagnose causes and mark as inferred.
8. Mobile & page experience
- Mobile-first: same content on mobile and desktop; responsive viewport set.
- HTTPS everywhere; no mixed content. No intrusive interstitials that block content.
- Tap targets, legible font sizes, no horizontal scroll.
Output format
| Severity | Area | Finding | Evidence (measured/inferred) | Fix |
Severity scale:
- P0 blocker — prevents crawl/index/render (noindex on key pages, robots block, 5xx, render-blocked content, broken canonical/hreflang).
- P1 major — meaningful ranking/UX impact (CWV failing, redirect chains, soft 404s, sitemap errors).
- P2 minor — hygiene (anchor text, shallow internal links, minor redirects).
- P3 nice-to-have.
End with: a short prioritized fix order, and explicitly note any area you could
not verify due to lack of access. Hand schema findings to seo-schema-builder,
content/architecture-of-content to seo-content-brief, and gate changes through
seo-quality-gates.
See references/checklist.md for an expanded checklist and verification commands.
What ships with it: 2 files
4.0 KB alongside SKILL.md
references/
- checklist.md2.5 KB
- contract.yaml1.6 KB