AEO Agent
Agent. Runs the check every week.
The weekly AI answer check is a job in Ai1, the AI agent platform by MyZone AI: a recipe that runs on a schedule. Every week the AEO Agent asks ChatGPT, Perplexity, Claude and Gemini about ten buyer questions you approved, and reports your AI visibility score, your share of voice, the sources the answers cite and what to fix first. It only reports.
Know where you stand
One score each week for whether AI answers recommend you.
See who they cite
The sites AI answers rely on, so you know where to show up.
Fix the right thing first
A prioritised list of what to change, every month.
At a glance
| Starts | A set time each week |
|---|---|
| Schedule | Every week, at a time you choose |
| Needs from you | About ten buyer questions you approve, and your connection to each AI engine |
| You get | A weekly dashboard and a monthly trend report with a fix list |
| If it can't finish | A passing problem, such as a brief outage, gets a limited number of retries. If the run still cannot finish, it goes to the job's owner instead of looping. |
| Agent | AEO Agent |
| Engines | ChatGPT, Perplexity, Claude and Gemini, each through your own connection |
| Changes | Nothing. It only reports |
More buyers now ask an AI assistant before they search. If the answer names your competitor and not you, you never hear about it. AI answers change from week to week, so a single check is not enough.
How this differs from the AEO Agent on its own: the job runs the same approved questions every week, so the trend means something.
How it works
A simplified view. It only reports; you decide what to change.

You confirm the buyer questions before tracking starts.
Runs every question through each engine, the same way every week.
Whether you are named, how often, and which sources are cited.
A branded weekly dashboard, and a monthly trend report with a competitor gap map and a fix list.
What you get every week
Example with a fictional business. Names and figures are invented to show the kind of check you get.
What it was asked: Do AI answer engines recommend us when buyers ask about what we do?
| Questions asked | Ten buyer questions, such as “Which cooler keeps ice longest for a weekend trip?”, asked in ChatGPT, Perplexity and Claude. |
|---|---|
| AI visibility score | 34 out of 100, up from 29 last week. Fernbrook Outdoor is named in 9 of 30 answers. |
| Share of voice | Fernbrook Outdoor 18%, next to Quillstone Coolers at 41% and Tallowmere Gear at 27%. |
| Word for word | Perplexity now names Fernbrook Outdoor for “best cooler for kayak trips”. ChatGPT still names only Quillstone Coolers. |
| Who they cite | Two review sites and one forum are behind most answers. Fernbrook Outdoor is not on either review site. |
| Fix first | Get the cooler range onto the two review sites, and add a plain answer page for the ice-time question. |
Built-in guardrails
Here is how the weekly check keeps you in control.
Better together
The agents, skills and recipes behind this job.
Agent. Runs the check every week.
Recipe. Adds the articles that help AI answers name you.
Agent. Its weekly digest covers classic search.
Job. The week's score can show up in your brief.
Common questions
ChatGPT, Perplexity, Claude and Gemini. It works with each service you connect, and their usage costs apply.
The same questions are asked in each engine every week, so you can see where you are named and where a competitor is named instead.
You do. The AEO Agent suggests about ten buyer questions, the kind your customers ask before they buy, and you confirm them before tracking starts.
Keeping the same questions each week makes the trend meaningful.
No. It only reports.
The monthly report includes a prioritised list of fixes for your team, such as getting listed on the review sites the AI engines cite most.
It takes a new baseline, so the trend stays honest.
Otherwise the score would move for reasons that have nothing to do with how often you are named.
We will run the AI answer check on your own buyer questions and walk you through the first result.