securing-kubernetes-on-cloud

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

• Third-party content exposure detected (high risk: 0.75). The workflow includes runtime fetching of outsider-authored free text from public web content (e.g., kubectl apply -f https://raw.githubusercontent.com/aquasecurity/kube-bench/main/job-eks.yaml), which the agent would ingest as readable YAML/command text into its LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The workflow runs live installs that fetch and apply remote manifests at runtime (helm repo add https://falcosecurity.github.io/charts and kubectl apply -f https://raw.githubusercontent.com/aquasecurity/kube-bench/main/job-eks.yaml), which executes remote code/manifests in-cluster and are required steps, so they are runtime external dependencies that can control behavior.