Your monitoring activity across all scans.
What you spend screen time on, regardless of which browser or app it's in. The basis for future pattern insights.
Scans per day. Red marks days with infractions.
Most recent scans reported by the monitor.
Create your own: a name, an optional word/site list (step 1 — fast on-screen text match), and an optional AI instruction (step 2). When you create it we'll offer to sharpen your AI instruction first.
Apps and content to ignore — still logged, but never counted as an infraction or blocked. Ships with a few hundred defaults (AI assistants, editors, terminals, dev and reference sites). Each one is a hole in the blocker, so switch off anything you'd rather still be caught on.
Install StarBlocker, sign in, and it shows up here.
Checking for the latest build…
Download StarBlockerFour steps, then it runs on its own.
.dmg, then double-click StarBlocker.pkg.
It asks for your administrator password — it installs a background
service that starts at login.Once a rule is active you won't be able to quit the app. That's the point.
StarBlocker monitoring requires an active subscription. This is a stub — no payment is collected yet. Flip the switch to simulate subscribing.
Your StarBlocker account details.
What you do to earn a change while your rules are running: switching a rule off, editing one, deleting one, adding to the whitelist, or changing this setting. Pick it in a calm moment — you'll meet it in a bad one.
When none of your rules are active — you're somewhere none of them apply — none of this is asked for and everything is free to change.
Which model classifies each scan. Prices are what the provider charges per million tokens; the projection is what that works out to for your screens, so you can price a switch before making one.
Runs the same image through each selected model with the same prompt, and reports what each one decided, what it cost, and how long it took. Nothing here is logged as a scan — it won't touch your infraction meter or your cost history.
What each model actually costs you per scan, measured from real usage. Average per scan is the comparable number — total spend mostly reflects how many scans each model happened to run.
Most recent scans, including which model classified each one.