Describe the change you care about, and Starizon checks the page on a schedule. Watch a number, keywords, a page area or a condition that needs AI to judge it.

Describe the page and condition

Open the toolbar’s Tools menu and choose Schedule → Create → Page monitor. The panel closes and puts /monitor in the chat input. Describe the page, the area to watch, the condition, the interval and the notification channel, then send the request. Answer any follow-up questions in the conversation.

You can also type /monitor in the AI Chat input, choose the command and add your request. You can create a monitor without switching to Agent S6.

/monitor Check this course catalog every morning. Notify me when a new beginner design course taught in English is listed for under $100.

/monitor Check the stock message on this product page every hour. Send a browser notification when it becomes available. Stop after the first match.

Use the actual page URL, and name the price, availability message or list you care about. “Let me know when this changes” gives Starizon less to check against.

Review the monitor

Review the generated setup before confirming. Then open the monitor in Schedule → Edit and check Watched area, Change condition, Actual rule, the schedule and Notify via. New monitors start with browser notifications; choose another configured channel here.

For conditions it can check directly, Starizon uses a number or keyword rule. For the rest, AI decides whether the current state or a change matches your request. Read the generated rule and confirm it says what you meant.

Some rules check what is on the page now. Others compare it with an earlier reading or a saved reference reading, called a baseline. The first successful check may set that baseline. If the condition already matches when you create the monitor, choose whether to notify you right away or record the starting state without an alert.

The monitor checks the page at the times you set, so a change that comes and goes between checks can slip past it.

Avoid repeated alerts

Alert stability sets how many checks in a row must meet your condition before an alert. It also sets how many checks must show that the condition has cleared before another alert can be sent. A number rule may offer a recovery threshold: the value it must reach before the monitor can alert again.

A condition that keeps matching may stay quiet after the first alert. The monitor history explains why an alert was skipped, such as waiting for another check or for the minimum time between alerts to pass.

Turn on Stop after the first match when one occurrence is enough. Checks pause after the match, and notifications already created can still go out.

Read the history

The monitor’s history lists successful checks, unchanged pages and failures. It tracks the page condition and the notification delivery as separate entries. A condition can match while its email is still pending or has failed, for example.

Could not judge and Check failed leave the condition unknown. Read the explanation, fix the page or account problem and run the check again when it makes sense. After a failed judgment, the monitor keeps the previous baseline.

With Keep check screenshots on, the monitor stores recent captures of the visible page area on your device. Open them from history to see what each check saw.

Pages that need your browser session

Each check needs the page to be reachable. Keep the required account signed in, and check the browser-window mode if the page behaves differently in a hidden window. A verification screen or a changed layout can stop the check from reading the watched area.

If the monitor cannot find the area, edit Watched area to describe where it is now. Repeated failures put the monitor into Needs attention. Resolve the issue before you resume it.

Chrome must stay running. Scheduled tasks covers missed checks and plan limits, and Notifications covers delivery.