kubernetes-patterns
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The workflow involves deploying privileged DaemonSets and MutatingAdmissionWebhooks. These are high-privilege operations required for eBPF kernel-level tracing and automated sidecar injection within a Kubernetes cluster.
- [REMOTE_CODE_EXECUTION]: Skill instructions describe a process for writing and compiling kernel-level C code for eBPF kprobes at runtime using BCC or Cilium toolchains. This behavior is documented as a core feature for zero-instrumentation observability.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it ingests untrusted cluster data (RBAC configurations, Prometheus scrape configs) to generate executable manifests and BPF code. No specific malicious patterns were found, but the capability exists by design for manifest generation.
Audit Metadata