How it works
Three steps, about ten minutes to set up. No app for your child to learn and nothing for them to check in with.
Install the agent on your child's device
One command on their Mac. It runs quietly in the background and takes a screenshot at an interval you choose.
AI reviews the screen activity
Each screenshot is classified against six risk categories, from grooming to gambling. Ordinary activity is ignored.
You get an email alert within minutes
When something unsafe appears, an alert lands in your inbox with a short summary. No alert means nothing was flagged.
What you get
Built around one job: tell you when it matters, and stay quiet when it does not.
Six risk categories
Explicit content, grooming, cyberbullying, self-harm, violence, and gambling. Each finding comes with a severity so you know what to open first.
Email, not another app
Alerts arrive in the inbox you already check. There is no second app to install and no feed to doomscroll.
Summaries, not transcripts
The AI describes what it saw in a sentence. It does not quote your child's private messages or send you the screenshot.
Ten-minute debounce
The same category is not re-alerted for ten minutes, so one bad afternoon does not become forty emails.
Multiple kids, one dashboard
Install the agent on each device with its own name. Every alert is tagged with which device it came from.
Honest privacy terms
Your data is not sold, not shared, and not used to train anything. Read the privacy page, it is one page long on purpose.
Simple pricing
One price covers every child and every device in your family.
Monthly
Month to month. Cancel any time, no phone call required.
- Unlimited child devices
- Email alerts within minutes
- Parent dashboard feed
- All six risk categories
Yearly
Works out to $6.58 a month. Bark Premium is $99 a year for screen time, location and more. If you just want the alerts, this is that part for $79.
- Unlimited child devices
- Email alerts within minutes
- Parent dashboard feed
- All six risk categories
Questions parents ask
Is my child's data sold?
No. Your data is never sold, rented, or shared with advertisers or data brokers. It is used for one thing, sending you alerts.
Will my child know it is running?
The agent shows up as a background process, and we recommend telling your child it is there. LumaGuard is built for a conversation, not surveillance.
What exactly does the AI see?
Screenshots are classified and reduced to a short summary. Summaries never repeat your child's private messages word for word, and raw screenshots are not emailed to you or kept.
How fast do alerts arrive?
In practice, a few minutes. The agent reviews on the interval you set (default 2 minutes) and alerts go out as soon as something non-safe is classified.
Can I cancel?
Yes, any time, from the email receipt or by replying to any alert. No phone call, no retention script.
Know before it becomes a story
Start with early access. Setup takes about ten minutes and the first alert email shows you exactly what a finding looks like.
Get early access