🌐 GEO & AEO AI Agent Optimization Skill

This skill defines deterministic rules, schemas, and SOPs for optimizing web content, technical documentation, and AI agent output for Generative Engine Optimization (GEO) and Answer Engine Optimization (AEO).

🎯 1. Core Principles of GEO/AEO vs Traditional SEO

📐 2. Deterministic Content Formatting Rules (For AI Agents)

1. Inverted Pyramid Answer Block (First 60 Words):
   • Immediately below any major heading (<h2>, <h3>), place a 40–55 word direct, declarative answer.
   • Prohibit pleasantries ("In this article...", "It is important to remember").
   • Formula: [Entity/Topic] is [Category/Definition] that [Primary Function/Mechanism] by [Concrete Differentiator].
2. Structured Table Extraction (Markup Rule):
   • Always format multi-item comparisons (features, pricing, benchmarks) into native HTML <table> blocks with strict <th> and <td> tags.
   • LLM crawlers prioritize tabular data over narrative prose for comparative queries.
3. Information Gain & Primary Statistics:
   • Every article must contain at least 2 proprietary or specific numerical findings (e.g., "Based on an evaluation of 500+ endpoints...").
   • LLMs favor citing primary sources with novel metrics rather than duplicate paraphrases.
4. Entity Bold Formatting in Lists:
   • In bullet points, always start with a bold entity label (e.g., **1. Protocol Latency:** ...).

🏷️ 3. JSON-LD Entity Graph Schema Template

Every optimized page must inject unambiguous Schema.org entity definitions linking to external knowledge graphs (Wikidata / Crunchbase):

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Organization",
      "@id": "https://example.com/#organization",
      "name": "BrandName",
      "url": "https://example.com",
      "sameAs": [
        "https://www.wikidata.org/wiki/Q...",
        "https://www.crunchbase.com/organization/..."
      ]
    },
    {
      "@type": "TechArticle",
      "@id": "https://example.com/post-slug/#article",
      "isPartOf": { "@id": "https://example.com/#website" },
      "headline": "Exact Target Topic Headline",
      "about": [
        {
          "@type": "Thing",
          "name": "Primary Entity",
          "sameAs": "https://en.wikipedia.org/wiki/..."
        }
      ],
      "author": {
        "@type": "Person",
        "name": "Yevhen Shaforostov",
        "jobTitle": "AI Product Manager & Full-Stack AI Engineer",
        "sameAs": "https://github.com/yevhens-hue"
      }
    }
  ]
}
</script>

✅ 4. AI Agent Quality & Validation Gate (Pass/Fail Checklist)

Before publishing or accepting an agent output, run this 5-point verification:

1. Inverted Answer Block: Does the H2 have an immediate 40–55 word direct definition?
2. Entity Ambiguity: Are all brand and technical terms linked to unambiguous schema definitions?
3. Information Gain: Does the text contain at least one verifiable quantitative statistic or benchmark?
4. Tabular Comparison: Is comparative data presented in a clean <table> instead of a paragraph?
5. Zero Fluff: Are all filler introductions removed?