eliasoulkadi/shokunin/seo-geo

SEO & GEO Architect

Optimize for traditional search AND AI-powered engines. Based on Google Search Central, Moz, and GEO research (2025-2026).

Sub-Commands

Workflow

Step 1: Run SEO audit

Check: title tags (< 60 chars, keyword first), meta descriptions (< 155 chars, value prop), H1 (one per page, keyword), img alt text, OG tags, canonical, robots.txt, sitemap.xml, Core Web Vitals.

Score: 90-100% good. 70-89% fix flagged. < 70% structural issues first.

Step 2: Run GEO audit

Check: llms.txt presence, ai-plugin.json, FAQ/HowTo schema, entity clarity, brand mention structure, answer format (direct answer in first 2 paragraphs).

Step 3: Prioritized fixes

Step 4: Structured data (must implement)

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Brand Name",
  "description": "One sentence. What you do.",
  "url": "https://example.com",
  "logo": "https://example.com/logo.png",
  "sameAs": ["https://twitter.com/handle", "https://github.com/handle"]
}

Also: Article, FAQPage, HowTo, Product, BreadcrumbList, LocalBusiness (if applicable).

Step 5: llms.txt

# Brand Name
> One-line description.
 
## Core pages
- https://example.com/: Home
- https://example.com/about: About
 
## Docs
- https://docs.example.com/api: API reference
 
## Blog
- https://example.com/blog/post: Title

Place at /.well-known/llms.txt. This is how AI crawlers (GPTBot, Claude, Gemini) discover your site.

Step 6: GEO optimization

Production Checklist

• [ ] Technical SEO > 90%
• [ ] GEO audit passes (llms.txt, schema, entity clarity)
• [ ] Core Web Vitals pass (LCP < 2.5s, INP < 200ms, CLS < 0.1)
• [ ] Structured data: Organization + page-specific types
• [ ] Sitemap.xml submitted to GSC
• [ ] robots.txt correct (not blocking anything important)
• [ ] hreflang configured (if multilingual)
• [ ] Canonical URLs on every page
• [ ] Open Graph + Twitter Cards on every page
• [ ] llms.txt published
• [ ] GSC monitoring active

Anti-Patterns

Technical SEO

Core Web Vitals Thresholds

Measure with: Lighthouse, PageSpeed Insights, Web Vitals extension, CrUX dashboard.

Crawl Budget Optimization

User-agent: *
Allow: /
Disallow: /api/
Disallow: /*?*
Disallow: /search
Sitemap: https://example.com/sitemap.xml

Rules:

• Block faceted navigation URLs with ? parameters
• Block internal search pages (no SEO value, consumes budget)
• Keep sitemap under 50,000 URLs or split into sitemap index
• Set lastmod accurately (Google prioritizes recently-changed URLs)
• Use <priority> sparingly (Google largely ignores it)

Hreflang for Multilingual Sites

<link rel="alternate" hreflang="en" href="https://example.com/en/page" />
<link rel="alternate" hreflang="es" href="https://example.com/es/pagina" />
<link rel="alternate" hreflang="x-default" href="https://example.com/" />

Rules:

• Every language must link to every other language (bidirectional)
• x-default for language selector page or auto-redirect
• Use ISO 639-1 codes (en, es, fr, de, ja, zh)
• Verify with Google Search Console International Targeting report

Schema.org Nesting Patterns

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Title",
  "author": { "@type": "Person", "name": "Author" },
  "publisher": { "@type": "Organization", "name": "Site", "logo": { "@type": "ImageObject", "url": "logo.png" } },
  "mainEntityOfPage": { "@type": "WebPage", "@id": "https://example.com/article" }
}

Nest entities, don't flatten. Google prefers deeply-nested schema over flat arrays.

GEO Optimization (Generative Engine)

For AI search engines (ChatGPT, Gemini, Perplexity, Google AI Overviews):

1. llms.txt at /.well-known/llms.txt with structured content
2. Entity clarity: every page must answer "who, what, where, when" in first 100 words
3. Citability: include statistics with source links (AI Overviews prefer cited content)
4. Direct answers: format answers as Q&A pairs (AI extracts these verbatim)
5. Long-form depth: AI favors pages > 1,500 words with clear section headers
6. Markdown structure: AI parses ## headers better than <h2> tags

Sources

• Google Search Central
• Google E-E-A-T guidelines
• schema.org documentation
• Moz Beginner's Guide to SEO
• GEO research papers (2025-2026)
• Perplexity Publisher Guidelines
• OpenAI GPT crawler docs

Error Handling

Checklist

• [ ] Skill loads without errors in the AI agent
• [ ] YAML frontmatter is valid (description, compatibility, audience)
• [ ] Workflow section provides clear step-by-step instructions
• [ ] Error handling section covers common failure modes
• [ ] All referenced files (references/, scripts/, assets/) exist
• [ ] Skill triggers correctly for intended use cases
• [ ] No broken links or missing resources