Today we're launching PC Health — a new capability in K12Panel that continuously measures the condition of your managed Windows devices, tells you when something crosses a line you care about, and, when you want it to, fixes the problem for you and verifies it's actually resolved before anyone files a ticket.
If you want the full concept walkthrough, start with About PC Health and Self-Healing. Otherwise, here's everything that's new.
At the heart of PC Health is the health check — a small measurement taken on a device that answers one question with one number. K12Panel ships with built-in checks for the basics:
And you're not limited to those. If a short PowerShell command can measure it, you can turn it into a check — a service's running state, free memory, certificate days-to-expiry, last backup age, whatever matters to you. Writing one is simple: your script computes a number (or throws if it can't), and K12Panel handles the rest. See Writing Custom Health Checks for the contract and copy-paste examples.
A check only measures — what counts as a warning or a critical is decided in a policy. This separation is deliberate: the same "disk free" check can be strict for a lab server and relaxed for a student laptop, just by applying a different policy.
Policies let you:
Every monitored device gets a Health tab showing live gauges and stoplights for each check, its current status (Healthy / Warning / Critical / Unknown), and a clear list of open issues — each on its own severity-banded card so nothing gets lost. If an issue is expected or being handled elsewhere, you can mute it for 7 days, 30 days, or indefinitely, and unmute it any time. Muting silences the alert (and pauses automation) without hiding the record.
When an issue crosses the line, K12Panel raises a PC Health Alert and can email the administrators who subscribe — so your team learns about the almost-full disks and dead services from one place, proactively, instead of from a ticket. Muted and actively-healing issues stay quiet.
Here's where it gets powerful. PC Health borrows a pattern from the security-operations world called SOAR — detect → decide → act → verify — and applies it to ordinary device health. When a check fails, K12Panel can run a remediation (a fix), then re-run the check to confirm the problem is actually gone.
The most important idea: the check is the source of truth, not the fix. A cleanup script can run cleanly and still not free enough space — so K12Panel never trusts a fix's own "success." It re-measures, and the issue only clears when the check passes. You'll see both outcomes in the log: what the fix reported, and whether the check agreed.
Automation that runs commands on devices has to be careful, so PC Health has guardrails built in:
Built a great policy — the right checks, thresholds, and fixes — for one school? Export it as a single bundle (policy, custom checks, remediations, and the wiring between them) and import it into another organization. Imported items land inactive pending review, so nothing runs until you say so.
PC Health runs on the Windows agent you already have. Checks and fixes ride the agent's normal check-in, so adding or editing one takes effect on the next check-in — there's nothing to deploy to the device and no agent update to wait for.
You don't have to use the automation to get value, and you can add your own checks and fixes whenever you like. Start simple and grow into it.