GEO/SEO Analysis & Audit Skill
This skill defines the technical methodology and analysis workflow for auditing websites against Generative Engine Optimization (GEO) and AI-native search engine (ChatGPT, Perplexity, Gemini, Claude, Copilot, Google AI Overviews) ranking behaviors.
🎯 Core Objective
Audit any target domain's visibility, technical crawlability, and semantic authority under conversational AI models, producing a standardized, high-value assessment report with actionable optimization roadmaps tailored to the type of web application (Agent-First vs. Human-First).
📊 Differentiated Scoring Scorecard (100 Points Total)
To cater to the distinct traversal paths of autonomous agents versus human-oriented conversational search, the audit uses a tailored metric weighting scheme:
| Metric | Max Points | Agent-First Heuristic | Human-First Heuristic |
|---|---|---|---|
| 1. Technical Accessibility | 15 pts | - AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.) permitted in robots.txt. |
- Standard search engine crawlers (Googlebot, Bingbot, etc.) permitted in robots.txt. |
| 2. Structured Schema Quality | 10 pts | - Validity of JSON-LD schemas mapped to agent capabilities or software specs. | - Presence and validity of schemas like Product, FAQPage, Organization, HowTo. |
| 3. Content Extractability | 15 pts | - Section headings, markdown structures, bullet points, and data grids. | - Structured comparison layout, QA headers, H1/H2/H3 hierarchies. |
| 4. Answer Density (BLUF) | 10 pts | - Concise summary blocks for machine ingestion. | - High-density 40-70 word answer blocks ("Bottom Line Up Front" formatting). |
| 5. Citations & Empirical Data | 15 pts | - Hard data points, numeric parameters, and outbound documentation references. | - Specific numbers, percentages, trust factors, and reliable source linking. |
| 6. Expertise & Attribution (EEAT) | 10 pts | - Developer/Publisher credentials, software repositories, and compliance/trust signals. | - Verified author credentials, profile bylines, customer review indicators. |
| 7. Off-Page Entity Footprint | 15 pts | - Hyperlinked references in code repositories, developer registries, or package indexes. | - External citations on mainstream wikis, reddit homebrews, forum indices. |
| 8. Machine-Readable Discovery | 10 pts | - Presence and compliance of /llms.txt and /skill.md at domain root. |
- Presence of standard /sitemap.xml and OpenGraph/Twitter social meta tags. |
🛠️ Step-by-Step Auditing Workflow
Step 1: Local & Automated Crawl Check
Run the automated python script geo_audit.py to calculate the baseline score and scan technical endpoints:
# Audit a B2B SaaS or E-commerce site (Human-first)
python3 geo_audit.py soldy.ai --type human --prompt "AI video generator"
# Audit an agent network app or dev protocol (Agent-first)
python3 geo_audit.py emergence.science --type agent --prompt "openclaw bounty market"
Step 2: E2E Search Engine Indexation Verification (Optional)
To query live index footprints and rank tracking on Tavily, Brave Search, Google, and Bing, append the --e2e flag:
# Requires local CLI tools: `bx` (Brave Search) and `tvly` (Tavily Search)
python3 geo_audit.py emergence.science --type agent --prompt "openclaw bounty market" --e2e
Step 3: Content Optimization Scan
- Review target landing pages to check if answers are buried in non-indexable structures (dynamic client-side React code, login walls).
- Count word length of key introductory paragraphs to check if they fit the 40-70 word extraction window.
Step 4: Sourcing & EEAT Validation
Verify if claims rely on subjective marketing adjectives instead of verifiable scientific and quantitative facts.
📋 Standard Reporting Structure & Examples
Every generated report must include:
- Executive Summary: Core highlights and final benchmark score.
- Current State Assessment: Breakdown of the 8 metrics with specific website screenshots or parsed snippets.
- Competitor Citation Gap Analysis: Why competitors get cited in ChatGPT/Perplexity for target keywords while the client is omitted.
- Actionable Roadmap: Clear 30-60-90 day recommendations (Technical fixes, structural copy rewrite, schema implementation, agentic files integration).
Live Example & Downloadable Report
- Live Interactive Tool & Report Demo: Visit Emergence Science SEO-GEO Audit Tool to view live report examples and score calculations.
- Reference Document: An example of a generated English audit report is bundled with this skill at examples/emergence_science_analysis_en.md for direct reference and templates.