advanced-kubernetes

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow for this operator reads outsider-authored free text from Kubernetes custom resource fields (e.g., .spec.image, .spec.resources.*, .spec.env[*].name/value) submitted by users into MyApp objects, which are then ingested by the controller and admission webhook logic during Reconcile/Validate*/Default.

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

• Potentially malicious external URL detected (high risk: 0.90). scripts/setup-operator-dev.sh performs a runtime kubectl apply -f of the remote manifest https://github.com/cert-manager/cert-manager/releases/download/v1.13.2/cert-manager.yaml, which fetches and applies remote resources (executing remote configuration) and is a documented dependency for webhooks/certificates in the skill.