Every check maps to a real behavior of large language models when they crawl, extract and cite content.
Checks the foundational HTML signals that LLMs use to understand page identity: title quality, heading hierarchy, semantic tags, canonical URL, Open Graph, language attributes, and hreflang.
Title tag
H1 / H2 / H3
Canonical URL
Open Graph
Semantic HTML
Lang / Hreflang
Read more
The highest-weighted pillar. Measures how well an LLM can extract, chunk, and cite your content: schema markup, paragraph length, lists, definitional patterns, dates, internal links, and breadcrumbs.
JSON-LD Schema
Paragraph length
Lists
Date signals
Internal links
Breadcrumb
Read more
Evaluates how clear and readable the content is for both humans and AI models. Checks average sentence length, heading density, and Flesch reading ease (skipped for Greek-language content).
Sentence length
Heading density
Flesch score
Greek-aware
Read more
Assesses the trust signals LLMs use to evaluate source credibility: About and Contact page links, author attribution, social profiles, image alt text, meta robots settings, and robots.txt AI crawler access.
About / Contact
Author
Social profiles
Image alt text
Meta robots
Robots.txt
Read more