Server / Host
A running server carries risks that code or image scans alone cannot see: the packages actually installed, the deployed infrastructure settings, and the runtime events happening right now. When you connect a server (host) over SSH, you inspect the running server itself, covering everything from installed package vulnerabilities to infrastructure settings and runtime events within a single service.
What you can do with this type
| Item | What it does |
|---|---|
| OS / package vulnerabilities | Collects packages installed on the server and checks for CVEs |
| Infrastructure checks | Security settings of running clusters/nodes (K8s/CIS/hardening) |
| Runtime monitoring | Installs a collector to receive runtime security events centrally |
You can enable only the items you want, and change them at any time after registration.
What you need
- Server SSH access details (host, port, user, password or key)
- For a server inside a private network, use multi-hop SSH through a gateway or a VPN connection.
- To inspect Kubernetes, you need a kubeconfig (or a service account token).
SSH/kubeconfig credentials are stored encrypted and shown masked on screen. Stored values are reused without re-entry.
Registration steps
- On the [Services] page, click Register service and select Server / Host.
- Enter the SSH access details. For a private network, add a gateway (hop).
- If you use Kubernetes checks, enter the kubeconfig in the dedicated K8s authentication step.
- Select the items to run.
- OS / package vulnerabilities: CVEs of installed packages
- Infrastructure checks: K8s / CIS / container hardening
- Runtime monitoring: runtime event collection
- Once registered, the selected items run.

Connecting to a private cluster
A private Kubernetes cluster behind a gateway is connected by opening a local TCP tunnel over the SSH chain. You can also connect via a VPN (WireGuard) profile. See Plans and feature scope for detailed conditions.