<aside>
TL;DR: Lighthouse’s new Agentic Browsing category evaluates how prepared your site is for AI agents
</aside>
Google Lighthouse now offers an **Agentic Browsing** category that audits how well a site is constructed for navigation, understanding, and interaction with AI agents acting on behalf of users.
The audit includes three main categories:
Instead of using a 100-point scoring system, the report displays a ratio score showing how many agentic readiness checks your site passes.
Note: This report is not accessible from the standard version of Chrome. Marketers must use Lighthouse within Chrome Canary (the experimental version for developers and early adopters) to see the Agentic Browsing category.
This report can serve as a Google-backed basis for recommendations and help search marketers justify resources from stakeholders; particularly for traditionally overlooked aspects of your digital presence (e.g., accessibility, future-proofing).
With regard to the confusion over llms.txt, best practices may be getting clearer: For generative AI search, Google says llms.txt is not required. This suggests that llms.txt is more important if your website has elements that are specific to AI agents (which is what Lighthouse’s Agentic Browsing report is all about).
<aside>
Use the following tracking link or your exclusive one:
For these topics, use the Tracking Parameter: Shared ID and add the exact URL of the publication. Click Save and copy the generated link.
You get $300 per Semrush One sale and $10 per trial activation.
</aside>
<aside>
Google’s tools are an important part of any visibility stack, but they only tell you about your agentic readiness and success on Google surfaces. Monitor your presence on every platform your customers use with Semrush One. Track your visibility across ChatGPT, Perplexity, Google AI Mode, and Bing, plus use the Site Audit tool to catch pages blocked from AI search.
Google is actively building infrastructure to support AI agents. Recent developments include: