Agent Visualizer
A score tells you how you did. This shows you what happened: where an agent arrives on your page, where it reads, where it stops, and everything after that it never reaches.
It costs one page, the same as an Analyzer scan, because it is the same scan — and you are never charged for a page we cannot read. Every visualization links to the full Agent Analyzer report.
What you are watching
Seven steps, in the order an agent meets them: it arrives, reads the words out of the markup, separates your content from the navigation, works out what the page is, finds who wrote it, looks for the rest of the site, and finally tries to do something other than read.
Each step is impossible until the one before it works. That is the part a number hides. A page can have immaculate structured data and a perfect author block and still be a dead end, because the agent stopped two steps earlier and never saw any of it.
So anything after a stop is drawn as unreached rather than passing. It did not succeed; it was never tried.
The same idea, for software
This walks a page. An agent working inside an application walks 7 steps of its own — discovering what the software can do, proving it is acting for someone, attempting the task, and getting back out with something the person can check.
The agent journey map for software →