k8s-core

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Kubernetes sources that may contain malicious instructions designed to influence the agent's behavior.
  • Ingestion points: Data is ingested from the cluster environment via kubectl_get, kubectl_describe, and kubectl_logs (SKILL.md).
  • Boundary markers: The instructions do not provide delimiters or specific guidance to ensure the agent distinguishes between its core instructions and the content retrieved from resource logs or descriptions.
  • Capability inventory: The skill provides access to high-privilege tools such as delete_resource, kubectl_apply (capable of modifying cluster state), and node_management (capable of draining or cordoning nodes) (SKILL.md).
  • Sanitization: No explicit sanitization or content validation requirements are defined for data retrieved from the cluster before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 06:52 PM