Methodology

How AI readiness score is calculated

The score is a diagnostic indicator, not a ranking guarantee. It measures whether a site can be reliably interpreted by AI systems through public technical and semantic signals.

What we evaluate

Availability and baseline technical readability, including lightweight HTTP, browser-like, markdown negotiation, search-style and AI-style synthetic probes.
Structure and navigational clarity.
Trust/contact signals on public routes.
Machine-readable layer: company-profile.json, agenthub.json, llms.txt, structured data, canonical Agntbase hub profile, and optional A2A agent-card.json when a real agent-facing interface exists.
Agent-readiness and decision clarity signals.

What this score does not mean

It does not guarantee citations in ChatGPT or AI Overviews.
It does not replace full SEO analytics stack.
It is not absolute truth for every dynamic site architecture; synthetic user-agent probes show how Agntbase's server is treated, not verified crawler-IP access.

Correct interpretation: use the score as a practical readiness baseline and follow the recommended fixes in order of impact.

Related page: AI Site Check.