1---2name: geo-fundamentals3description: "Optimizes content for citation in AI answer engines. Use when improving AI search visibility or running a GEO audit."4allowed-tools: Read Glob Grep5compatibility: "Claude Code with the Cardume plugins (uses cardume-* commands, subagents or plugin paths); the guidance itself is portable."6metadata:7 source: https://skillmd.com/skills/jeffersonvalente/geo-fundamentals8---910# GEO Fundamentals1112> Optimization for AI-powered search engines.1314---1516## 1. What is GEO?1718**GEO** = Generative Engine Optimization1920| Goal | Platform |21|------|----------|22| Be cited in AI responses | ChatGPT, Claude, Perplexity, Gemini |2324### SEO vs GEO2526| Aspect | SEO | GEO |27|--------|-----|-----|28| Goal | #1 ranking | AI citations |29| Platform | Google | AI engines |30| Metrics | Rankings, CTR | Citation rate |31| Focus | Keywords | Entities, data |3233---3435## 2. AI Engine Landscape3637| Engine | Citation Style | Opportunity |38|--------|----------------|-------------|39| **Perplexity** | Numbered [1][2] | Highest citation rate |40| **ChatGPT** | Inline/footnotes | Custom GPTs |41| **Claude** | Contextual | Long-form content |42| **Gemini** | Sources section | SEO crossover |4344---4546## 3. Retrieval Factors4748How AI engines tend to select content to cite (engines do not publish weights; treat these as qualitative signals):4950| Factor | What helps |51|--------|------------|52| Semantic relevance | Directly answers the question in clear, self-contained passages |53| Authority signals | Named authors, credentials, reputable mentions and links |54| Freshness | Visible "last updated" dates, current facts |55| Crawlability | Server-rendered HTML the engine's crawler is allowed to fetch |56| Source diversity | Engines mix sources, so unique data and angles stand out |5758---5960## 4. Content That Gets Cited6162| Element | Why It Works |63|---------|--------------|64| **Original statistics** | Unique, citable data |65| **Expert quotes** | Authority transfer |66| **Clear definitions** | Easy to extract |67| **Step-by-step guides** | Actionable value |68| **Comparison tables** | Structured info |69| **FAQ sections** | Direct answers |7071---7273## 5. GEO Content Checklist7475### Content Elements7677- [ ] Question-based titles78- [ ] Summary/TL;DR at top79- [ ] Original data with sources80- [ ] Expert quotes (name, title)81- [ ] FAQ section (3-5 Q&A)82- [ ] Clear definitions83- [ ] "Last updated" timestamp84- [ ] Author with credentials8586### Technical Elements8788- [ ] Article schema with dates89- [ ] Person schema for author90- [ ] FAQ content in plain HTML (FAQPage markup is optional: Google no longer shows FAQ rich results)91- [ ] Fast loading (< 2.5s)92- [ ] Clean HTML structure9394---9596## 6. Entity Building9798| Action | Purpose |99|--------|---------|100| Google Knowledge Panel | Entity recognition |101| Wikipedia (if notable) | Authority source |102| Consistent info across web | Entity consolidation |103| Industry mentions | Authority signals |104105---106107## 7. AI Crawler Access108109### Key AI User-Agents (robots.txt tokens)110111| Vendor | Token | Purpose |112|--------|-------|---------|113| OpenAI | `GPTBot` | Crawls content that may be used for model training |114| OpenAI | `OAI-SearchBot` | Surfaces sites in ChatGPT search answers |115| OpenAI | `ChatGPT-User` | User-initiated fetches; robots.txt may not apply |116| Anthropic | `ClaudeBot` | Collects content that may be used for model training |117| Anthropic | `Claude-SearchBot` | Indexes content to improve Claude's search results |118| Anthropic | `Claude-User` | Fetches pages when a Claude user asks a question |119| Google | `Google-Extended` | Control token only (no separate crawler) for Gemini training/grounding; does not affect Google Search |120| Google | `Googlebot` | Google Search, including AI features in Search |121| Perplexity | `PerplexityBot` | Surfaces and links sites in Perplexity search results |122| Perplexity | `Perplexity-User` | User-initiated fetches; generally ignores robots.txt |123124> Vendors add and rename agents; verify against their docs before shipping robots.txt: [OpenAI](https://developers.openai.com/api/docs/bots), [Anthropic](https://support.claude.com/en/articles/8896518), [Google](https://developers.google.com/search/docs/crawling-indexing/google-common-crawlers), [Perplexity](https://docs.perplexity.ai/guides/bots).125126### Access Decision127128| Strategy | When |129|----------|------|130| Allow search + user agents | Want AI citations (`OAI-SearchBot`, `Claude-SearchBot`, `PerplexityBot`, user agents) |131| Block training agents only | Opt out of training (`GPTBot`, `ClaudeBot`, `Google-Extended`) while staying citable |132| Selective | Allow some, block others per vendor |133134---135136## 8. Measurement137138| Metric | How to Track |139|--------|--------------|140| AI citations | Manual monitoring |141| "According to [Brand]" mentions | Search in AI |142| Competitor citations | Compare share |143| AI-referred traffic | UTM parameters |144145---146147## 9. Anti-Patterns148149| ❌ Don't | ✅ Do |150|----------|-------|151| Publish without dates | Add timestamps |152| Vague attributions | Name sources |153| Skip author info | Show credentials |154| Thin content | Comprehensive coverage |155156---157158> **Remember:** AI cites content that's clear, authoritative, and easy to extract. Be the best answer.159160---161162## Script163164| Command | Purpose | Command |165|--------|---------|---------|166| `cardume-geo-checker` | GEO audit (AI citation readiness) | `cardume-geo-checker <project_path>` |167