Technical and visibility diagnostics investigating how AI search engines read, extract, and represent your web domain, with findings grounded strictly in available evidence.
CiteFrame is Ignacio Rotth’s commercial GEO auditing service. GEO-Copilot is the separate open-source developer tool.
The GEO Audit investigates how modern generative search systems—including OpenAI SearchGPT / ChatGPT, Perplexity, Google Gemini, and Claude—crawl, extract, index, and cite your web domain. Rather than relying on generic SEO checklists or speculative ranking tricks, every audit is anchored in observable crawler diagnostics, structured data analysis, and multi-provider prompt probing.
This service is designed for:
robots.txt directives, and semantic HTML are digestible by AI bot crawlers without client hydration failures.
/llms.txt) and interconnected Schema.org knowledge graphs to ground AI agent responses in verified facts.
The diagnostic follows a rigorous 5-pillar technical framework grounded in information-retrieval research:
Audit of robots.txt access for AI search bots (OAI-SearchBot, PerplexityBot, Claude-Web), server-side rendering (SSR/SSG) availability, HTTP header negotiation, and zero-JS static extraction baselines.
Validation of interconnected Schema.org JSON-LD @graph schemas, Wikidata entity grounding (sameAs), author attribution, and corporate entity disambiguation.
Assessment of observable factual density, quantitative claims, comparison tables, domain-specific terminology, and ambiguous promotional prose.
Structure analysis for Bottom Line Up Front (BLUF) lead paragraphs, atomic self-contained semantic units (200–500 tokens), clean comparison tables, and accessible descriptive image metadata.
Structured prompt probing across leading generative models to test brand recognition, factual recall accuracy, competitor comparative context, and presence or absence of citations.
To maintain clarity between open-source tools and commercial consulting, the boundaries are strictly defined:
@geo-copilot/cli).Clients receive structured, evidence-backed documentation designed for both executive decision-makers and engineering teams:
Full technical assessment of crawler accessibility, bot header responses, structured data errors, and model probe responses across generative engines.
Strategic breakdown of your domain's AI visibility posture, verified evidence availability states, competitor landscape, and high-level priorities.
Prioritized engineering roadmap categorized by change safety (Class A non-breaking files, Class B head tags, Class C structural route updates).
Tailored, copy-ready specifications for /llms.txt, connected JSON-LD @graph schemas, and robots.txt bot permissions.
In accordance with our strict evidence integrity doctrine, the audit adheres to factual boundaries:
Deliverables and findings depend strictly on evidence available during the audit period. Citation evidence can be UNMEASURED when AI models omit explicit web citations, and competitor evidence may be unavailable depending on query retrieval behavior.
Generative AI search engines are non-deterministic, proprietary systems operated by third-party providers (OpenAI, Google, Anthropic, Perplexity) with frequent algorithmic retraining. The GEO Audit provides diagnostic measurement and engineering remediation; it cannot guarantee specific citation frequency, mention inclusion, search rankings, or commercial outcomes.
Generated code blueprints and implementation artifacts require review and testing within the client's codebase before being deployed to production.
Choose your preferred channel to get started or discuss custom requirements: