k8s-security-policies

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a configuration guide and template repository for Kubernetes security. All identified components are documentation-focused and promote defense-in-depth security practices.
  • [RBAC_PATTERNS]: Provides standard RBAC patterns for read-only access, namespace administration, and service accounts. While some patterns include powerful permissions (e.g., wildcard verbs for a namespace admin), they are clearly labeled as patterns and accompanied by best practice warnings against using wildcards in production.
  • [NETWORK_POLICIES]: Includes templates for network segmentation, such as 'default-deny-all' and specific allow-lists for DNS and cross-namespace communication. It correctly suggests blocking access to the cloud metadata service (169.254.169.254).
  • [POD_SECURITY]: Correct implements Pod Security Standards (Privileged, Baseline, Restricted) via namespace labels and pod-level security contexts.
  • [TROUBLESHOOTING]: Includes benign 'kubectl' commands for verifying permissions and network configurations without any command injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:55 AM
Security Audit — agent-trust-hub — k8s-security-policies