pod-management

Fail

Audited by Snyk on Nov 21, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The skill is a standard kubectl reference, but it lists destructive commands (force-deleting pods and deleting all pods in a namespace) and a port-forward bound to 0.0.0.0 without any confirmation or warning, which is a careless-risk pattern rather than malicious intent.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). The required workflow runs kubectl logs (including -f, --previous, and label-based multi-pod aggregation) and reads pod events and describe output, so workload-generated log lines that can carry outsider-supplied input (e.g., request paths or user agents from a public-facing service) enter the agent's context without it selecting specific text.

MEDIUM W013: Attempt to modify system services in skill instructions.

  • Attempt to modify system services in skill instructions detected (low risk: 0.10). The skill exposes state-changing cluster operations (kubectl apply/create, pod deletion including forced and namespace-wide deletion, and debug/run with node overrides), but it gives no explicit evidence of unrecoverable loss of meaningful operational state, so gate 2 fails and the tier is low.

Issues (3)

E004
CRITICAL

Prompt injection detected in skill instructions.

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

W013
MEDIUM

Attempt to modify system services in skill instructions.

Audit Metadata
Risk Level
CRITICAL
Analyzed
Nov 21, 2026, 11:43 AM
Issues
3