lemuriaOS
seoseocontentanalyticsgrowth

Generative Engine Optimization Audit

Audit a brand's content and technical setup for AI citation readiness, then produce a prioritized GEO action plan with measurable citation improvement targets.

Context

Use this skill when onboarding a new GEO client or running a quarterly audit. It produces the baseline score and action plan that drives all downstream GEO work (structured data, crawlability, brand mentions, LLM visibility tracking).

Procedure

  1. Audit robots.txt for AI crawler permissions (GPTBot, ChatGPT-User, PerplexityBot, ClaudeBot, Google-Extended, Bingbot).
  2. Evaluate content structure for AI extractability: presence of TL;DR sections, answer capsules, clear H-tag hierarchy, definition blocks, and FAQ sections.
  3. Assess E-E-A-T signals: author bios with credentials, citations to primary sources, original data or research, and expertise claims backed by evidence.
  4. Check structured data implementation: FAQPage, HowTo, Article, Organization, and LocalBusiness schema where applicable.
  5. Check Bing indexation status (ChatGPT uses Bing for web search retrieval).
  6. Review presence on third-party citation sources: Reddit, Quora, G2, LinkedIn, Wikipedia, and industry publications known to appear in LLM training data.
  7. Score each dimension using weighted rubric and produce the GEO Readiness Score (0-100).
  8. Output prioritized action plan sorted by citation-lift potential vs. implementation effort.

Output Format

# GEO Audit Report: [Domain]

## GEO Readiness Score: [X/100]

### Dimension Scores
| Dimension | Score | Weight | Weighted |
|-----------|-------|--------|----------|
| Crawler access | /20 | 20% | |
| Content structure | /20 | 20% | |
| E-E-A-T signals | /20 | 20% | |
| Structured data | /15 | 15% | |
| Bing presence | /15 | 15% | |
| Third-party mentions | /10 | 10% | |

## Crawler Configuration Status
- GPTBot: [allowed/blocked/missing]
- ChatGPT-User: [allowed/blocked/missing]
- PerplexityBot: [allowed/blocked/missing]
- ClaudeBot: [allowed/blocked/missing]
- Google-Extended: [allowed/blocked/missing]
- Bingbot: [allowed/blocked/missing]

## Priority Actions
| # | Action | Effort | Citation Lift | Measurement Method |
|---|--------|--------|---------------|-------------------|
| 1 | | L/M/H | L/M/H | |

## Content Restructuring Recommendations
[Per-URL or per-topic recommendations with specific structural changes]

## Next Steps
- Handoff to: structured-data (if schema gaps found)
- Handoff to: crawlability (if crawler access issues)
- Handoff to: brand-mentions (if third-party presence is weak)
- Handoff to: llm-visibility (to begin citation tracking)

QA Rubric (scored)

  • Crawler audit accuracy (0-5): verified against actual robots.txt, not inferred.
  • Recommendation specificity (0-5): each action has owner, effort estimate, and measurement method.
  • Citation mechanism mapping (0-5): every recommendation tied to a documented AI behavior.
  • Prioritization logic (0-5): effort/impact ordering is defensible and non-arbitrary.

Examples (good/bad)

  • Good: "GPTBot is blocked in robots.txt. Recommendation: add Allow rule for GPTBot. Measurement: re-check robots.txt + monitor ChatGPT citation within 30 days."
  • Bad: "Improve your content to be more AI-friendly." (vague, no specific mechanism or measurement)

Variants

  • Quick audit variant: crawler access + structured data only (2-hour turnaround for prospect qualification).
  • Deep audit variant: full 8-step procedure with competitor benchmarking and quarterly tracking setup.