chaos-engineer

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The GitHub Actions workflow applies a remote Litmus operator manifest at runtime via kubectl apply -f https://litmuschaos.github.io/litmus/litmus-operator-v2.14.0.yaml which fetches and executes remote Kubernetes manifests (operator/controller code) required for the experiments, so this external URL directly executes remote code during skill runtime: https://litmuschaos.github.io/litmus/litmus-operator-v2.14.0.yaml

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs running commands that modify cluster and system state (kubectl apply/patch, starting services, installing tools, and invoking chaos-monkey to kill instances), which can change or disrupt the machine/environment and require elevated privileges or sensitive access.