Google just shared more information to help us get our websites ready for agents. There is a new report from Lighthouse that anybody can run. You do not need external software to run it. You can do it right from within your Chrome browser.
The report tells you whether your website is discoverable for AI agents, whether you have WebMCP integration set up, and something worth discussing: an evaluation of your LLMs.txt file!
How To Run The Agentic Web Report
Right now, if you try to get this report from within the standard version of Chrome, you probably will not find it. You can use it if you have Chrome Canary. This is the upcoming beta version of Chrome. Once you have that, you simply right-click and choose Inspect Page, then navigate to Lighthouse at the top. You will see a new category for “Agentic Browsing.”
Walk Through The New Lighthouse Agent Readiness Report With Me
I ran this on Google’s own page about the new Lighthouse report for agentic browsing, and it turns out that Google’s own documentation has issues that might hinder agents! The new report does not give a score out of 100, but rather a ratio showing how many agentic readiness checks your site passes.
3 Things To Watch For Agentic Readiness
Here are the topics I have been discussing with my clients regarding this new shift.
1. AI Accessibility And The Accessibility Tree
Agents can look at your pages in three ways:
- Vision (screenshots).
- HTML.
- The accessibility tree.
The accessibility tree was originally meant for screen readers, but it actually tells an agent where the buttons are and what the important elements are. If your accessibility tree is not well-formed, agents will struggle to use your site. I think that making our pages agent-friendly will eventually be a ranking factor in terms of whether agents recommend your page.
2. Understanding WebMCP
WebMCP is a proposed web standard to help you build and expose structured tools for AI agents. Essentially, it is a way to teach agents how to use the functionality of your website.
There are two types: declarative and imperative. Declarative is simple code you wrap around a form, while imperative allows the agent to interact back and forth with your website. If you have tools on your site that people will use with their agents, WebMCP is going to be very important.
3. The LLMs.txt File
This sounds crazy because Google just put out documentation for ranking in the AI features of Search, saying you do not need an LLMs.txt file. But this report is not about Search; it is about agents using your website. The proposal is to use an LLMs.txt file (similar to robots.txt) to provide markdown information that helps agents understand your site at inference time. It allows you to give specific instructions to agents on what they are allowed to do and where they can find important information. You likely don’t need an LLMs.txt file unless you have elements that specifically are going to be used by agents.
LLMs.txt is for agents using your site, and not for Search reasons.
I would highly recommend setting aside some time to check your own site in Chrome Canary. Most of us do not need these files right this second, but we need to be aware of them as our websites start to become agentic.
More Resources:
Read Marie’s newsletter, AI News You Can Use. Subscribe now.
Featured Image: Roman Samborskyi/Shutterstock
