About AIVIS
AIVIS measures how AI search engines see a website, and tells you exactly what to fix so they can find, read and cite it.
What AIVIS does
Every scan crawls your homepage and key inner pages the way AI crawlers do (no JavaScript), runs 15 checks on access, structure, schema.org, freshness, citability, trust and speed, and asks 7 AI engines (ChatGPT, Claude, Gemini, Perplexity, Grok, Le Chat and more) what they actually know about your brand.
How we score
Deterministic checks are computed in code from the HTML, robots.txt, sitemap and llms.txt, so the same site gets the same score. Where a language model judges content, it picks from fixed categories and the score is calculated by code, not by the model. Every finding comes with the evidence we saw and a concrete fix.
Where the methodology comes from
The checks follow public sources: the robots.txt rules each AI vendor documents for its crawlers, schema.org, the llms.txt proposal, Google's E-E-A-T guidelines and the GEO study by Aggarwal et al. (2023) on what makes content cited by generative engines. Our own research pages publish aggregate results on more than a thousand scanned domains.
Independent
AIVIS is independent software. It is not affiliated with OpenAI, Anthropic, Google, Perplexity, xAI or Mistral, and no engine pays to be included or ranked.
Talk to us
Bugs, wrong results, billing, data requests and partnerships all go to one address, answered by the people who build the product.
Research · Changelog · Contact · deus+aivis@lumnika.com
Sources
On a benchmark of 10,000 queries, adding statistics, quotations and cited sources raised a page's visibility in generative-engine answers by up to 40%. Aggarwal et al., GEO: Generative Engine Optimization (KDD 2024)
AI search engines answer with retrieval-augmented generation, introduced by Lewis et al. in 2020: they fetch live web pages first, so a page they cannot read cannot be cited. Lewis et al., Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (NeurIPS 2020)