ams-seo-site-auditor
暂无简介
- License 许可
- MIT
- 最近更新
- 2026-11-07
- Stars
- ★ 0
- Forks
- 1
03 SKILL.md 正文
SEO Site Auditor
A systematic approach to technical SEO audits: crawling a site, checking indexability, metadata, structured data, performance, and content signals — then prioritizing fixes by impact and effort.
Overview
SEO audits fail when they're 200-line checklists with no priorities. This skill is a triage method: crawl → identify what's blocking indexing/ranking → verify the high-impact basics (titles, canonicals, sitemaps, speed, structured data) → prioritize by (traffic at stake × effort). Most sites' SEO problems are a handful of fixable technical issues, not mysterious algorithm behavior.
When to use
- Traffic drops or pages not appearing in search.
- Pre-launch SEO review of a new site or migration.
- Periodic technical health checks.
- Evaluating an agency/freelancer SEO proposal (know what good looks like).
Core concepts
- Crawlability. Can bots fetch and render pages? Check robots.txt, meta robots, HTTP status codes, redirect chains, and JS-rendering (view-source vs rendered DOM for key content).
- Indexability.
noindextags, canonical tags (self-referencing where appropriate), sitemap.xml accuracy, Search Console coverage report. - Metadata. Unique
<title>(50–60 chars, keyword-front-loaded naturally), meta descriptions (compelling, ~150 chars — they don't rank you but earn clicks), one H1 per page, logical heading hierarchy. - Structured data. JSON-LD schema.org (Article, Product, FAQ, BreadcrumbList, Organization) — validates in Rich Results Test; enables rich snippets.
- Core Web Vitals. LCP < 2.5s, INP < 200ms, CLS < 0.1 — measured via PageSpeed Insights / CrUX (field data beats lab data).
- Content signals. Thin/duplicate content, keyword cannibalization (multiple pages targeting the same query), internal linking depth (important pages ≤ 3 clicks from home).
- Migration risks. URL changes without 301s, lost metadata, staging
noindexleaking to production.
Practical workflow
1. Crawl.
# CLI crawl for status codes, titles, canonicals, noindex
npx broken-link-checker https://example.com --ordered
# or: screaming-frog-style desktop crawler for full audits
curl -sI https://example.com/robots.txt
curl -s https://example.com/sitemap.xml | head -50
Check: 200s where expected, no redirect chains (>2 hops), noindex only where intended, canonicals absolute and self-consistent.
2. Indexation check. Search Console: coverage errors, "discovered not indexed" (crawl budget/content quality signal), sitemap submission status. site:example.com for a rough indexed-page count.
3. Page-level audit (top 20 pages).
- Title unique and descriptive? Meta description present and click-worthy?
- H1 present, single, matching intent? Headings hierarchical?
- Canonical correct? Open Graph/Twitter cards render (test with validators)?
- JSON-LD valid? (Rich Results Test)
- Images: descriptive alt, compressed, modern formats, width/height set (CLS)?
4. Performance. PageSpeed Insights per template (not just homepage): LCP element identified, INP interactions, CLS sources. Fix images/fonts/JS bloat first — they're usually the cause.
5. Content. Identify: thin pages (< few hundred useful words), duplicates/near-duplicates, cannibalization (Search Console: two pages ranking for the same query — consolidate).
6. Prioritize. Score each finding: impact (traffic at stake) × effort (dev hours). Fix order: blocking (noindex leaks, 5xx) → high-impact quick wins (titles, canonicals) → structural (internal linking, speed) → content (consolidation, new content).
Common pitfalls
noindexon production. Staging configs leaking to prod — the single most expensive one-line SEO bug. Check after every deploy.- JS-rendered critical content. If the title, headings, or body text only exist after JS runs, crawlers may miss or delay them. SSR critical content.
- Canonical chaos. Canonicals pointing to wrong URLs, chains, or http/https mismatches. Every indexable page: self-referencing absolute canonical.
- Sitemap fiction. Sitemaps listing 404s, noindexed pages, or missing new content. Regenerate automatically; keep under 50k URLs per file.
- Title/description duplication. Same title on 500 product pages = missed clicks. Templates with unique variables (product name, category).
- Chasing scores over users. Perfect Lighthouse 100 with thin content ranks worse than useful content at 85. Technical SEO is table stakes; content wins.
- Ignoring Search Console. The only source of truth for how Google sees your site. Check coverage, enhancements, and manual actions regularly.
- Migration without redirects. Relaunching with new URLs and no 301 map = traffic cliff. Map every old URL → new (or closest equivalent) before launch.
- Keyword stuffing. 2010 called. Write for the query's intent; use the keyword naturally in title/H1/early body.
04 同仓库更多 Skills
web-scrape
Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open/fetch fails on JS-heavy pages, or when you need clean markdown/structured extraction from a site. Requires FIRECRAWL_API_KEY.
design-taste-frontend
tasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curated gallery sites (One Page Love, Awwwards Sites of the Day). Deals researched directions (idea, composition, type, palette, photographs, section blueprints) from a library, enforces a strict no-slop house style with a real-browser inspector, and runs a look-and-fix loop on screenshots. Use this whenever you build, design, redesign, restyle, polish or review a landing page, marketing site, portfolio, product page, one-pager, hero or any public-facing web page, in any stack (HTML/CSS, React, Next.js, Vue, Svelte, Astro, Tailwind) — also when the user says a page looks generic, templated, "AI-made", sloppy, or asks to make it look designed, premium, award-level or less like every other site. Not for dashboards or app screens.
test-driven-development
Use when implementing any feature or bugfix, before writing implementation code
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).