kubernetes-pentesting
Warn
Audited by Snyk on Nov 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow runs kubectl/curl commands against a target cluster and reads its RBAC, secrets, and pod data, which is target-owned rather than outsider-submitted, but pulling and inspecting a registry image (
docker history) exposes build-time text from an image the agent selects, which an outsider could have published to a public registry.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill is an offensive Kubernetes pentesting playbook that directs the agent to create privileged pods with host filesystem and nsenter access, bypass admission webhooks, and drop static pod manifests into node system directories, all of which compromise the machine's security state.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata