Containers (Navin Security)
Overview
Audit images, Dockerfiles, Compose files and Kubernetes manifests. Default posture is read-only and non-destructive: report evidence and a proposed fix, never exploit, extract data or change the target.
When it applies
- Triggers: docker, kubernetes.
- Families: infrastructure.
Controls covered
| Control | Level | Check |
|---|---|---|
| C4.05 | L1 | Containers: root user, images, docker.sock, capabilities |
| C4.06 | L1 | Kubernetes: RBAC, NetworkPolicy, securityContext, secrets |
Engines
- Registered scanners (use
navin.security.scanners, results arrive as normalized findings): trivy. - Other engines named by the catalog: checkov, kube_bench, navin_rules. Use one only when it is installed.
- A scanner that is not installed or did not run means the check was not done, not that it passed.
Method
- Dockerfile: non-root
USER, pinned base image, no secrets in build args or layers, minimal packages. - Compose and runtime: no
privileged, nodocker.sockmount, dropped capabilities, read-only root where possible. - Kubernetes: RBAC without wildcards, NetworkPolicy present,
securityContextset, no host namespaces, secrets not in plain env. - Use the registered scanners for image and manifest checks; confirm each hit against the manifest actually deployed.
- Fix as a diff on the manifest, with the setting and its safe value.
Evidence and status
potential: a rule or a suspicion with static evidence only.confirmed: a dataflow trace from a controlled source to the sink, or a dynamic observation on the running target.exploitable: a non-destructive reproduction showed the impact, in an authorized environment.- Every finding carries file and line or endpoint, the evidence, impact, a concrete fix and the control ids above.
Anti-patterns
- Scanning an image tag that is not the one deployed.
- Applying a manifest to a cluster to test it.
- Treating a clean scan as proof of security: say which controls were not assessed.
