Agent readiness

Find out whether an AI agent can actually use your page

A new class of AI browsers does not just read the web, it operates it: clicking buttons, filling forms, finishing a task someone asked for. We check whether one of them can reach your page, find the controls, and get through. No score — a verdict, the counts behind it, and one thing to fix.

yoursite.com/demo
🌐 yoursite.com/demo Operating…
"Work email" Work email
"Company" Company
Ready — an agent can reach, read and operate this page.
A different visitor

Being readable and being operable are not the same test.

A page can pass every crawlability and content check on this site and still be impossible to complete a task on.

A crawler needs to read

Reachable URL, HTML that contains the content, structure that says what the page is. It takes what it finds and leaves. Nothing on the page has to respond.

An agent needs to act

Labelled fields it can identify, controls it can operate without a mouse, states it can tell apart, and a confirmation it can recognise as success. One unlabelled input ends the task.

A run, step by step

Where the attempt actually ends

We report the specific step, not a score. “Book a demo” on a page that reads perfectly:

1 Reach the page and read its purpose Done
2 Locate the demo request form Done
3 Identify what each field wantsWork email and Company resolved from their labels. Done
4 Choose a team sizeA box styled to look like a dropdown, with no real control underneath it and no name anything can read. The agent cannot tell what the options are, or that it is a control at all. Stopped
5 Submit the form Not reached
6 Confirm the booking succeeded Not reached
The finding One control, four steps in. The person who asked their assistant to book a demo simply hears that it could not be done — and you never see the attempt in your analytics.
Common blockers

What tends to end the run

Most of these are also accessibility problems, so fixing them pays twice.

</>

Controls that only look like controls

A plain box styled to look like a dropdown or a button, rather than a real one. A person sees a control; an agent sees a decorated rectangle with nothing to operate and no options to choose from.

 ? 

Fields identified only by placeholder

Placeholder text disappears on focus and is not a label. Without one, nothing can say what the field is for.

Silent state changes

The page updates but announces nothing. The agent cannot tell whether it succeeded, failed, or is still waiting.

Interaction-gated steps

A hover-only menu, a drag control, or a step that needs a real pointer. Nothing automated will produce one.

1

unlabelled control is enough to end an otherwise complete task

2

problems fixed at once — agent access and screen-reader access are the same work

0

trace of the failed attempt in your analytics

Find out where the agent stops.

Run the check on any URL. You get the step-by-step attempt, the control that ended it, and what to change.

FAQ

Agents, answered

What is an AI agent actually doing on my page?

Driving a real browser to finish something a person asked for — book the demo, add to basket, get through the signup. It clicks, types and waits like a visitor would, which means it can be stopped by things a reader would never notice.

Does this change my score?

No, deliberately. Agentic browsing returns a verdict — Ready, Needs work, Blocked or No action found — and a count, never a number, and it is not one of the dimensions your AI Visibility Score is built from. Not every page should be operable, so scoring every page against it would be wrong.

My page came back “No action found”. Is that bad?

No. An article, a policy page, a piece of documentation has nothing to operate, and that is the correct answer for it. It is only worth investigating if you expected an action there — which usually means the control appears after an interaction the agent never got to make.

Is anyone actually browsing with agents yet?

The agents exist and are already in our crawler list — ChatGPT-User, Perplexity-User, Claude-User and GoogleAgent-Mariner among them. Volume is small today. What makes it worth doing early is that the work is the same work as accessibility, so it is not a bet on a timeline.

What actually stops one?

Three things, in order. Getting in at all — a login wall, a bot challenge, a hard block. Then whether the controls have names it can identify. Then whatever sits in the way: consent overlays, modals, interstitials, and whether each can be dismissed.

Does this actually submit my forms?

No. We walk up to the point of submission and report what an agent could identify and operate. You do not get test leads.

It says access is blocked. Where do I fix that?

On the crawlability side — that is where getting in is decided. A challenge that stops a crawler almost always stops an agent too, so fix it once and both improve.

How is this different from an accessibility audit?

They overlap heavily, which is the good news. An accessibility audit measures conformance against a standard. This runs one task end to end and tells you the step it stopped on.

Do I need to build an API for agents?

No. A well-built page is already the interface. Real form controls, proper labels and state changes the page announces get you most of the way.

Which plans include this?

Every paid plan.