k8s-cli
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
- [PERSISTENCE]: The skill documentation encourages the user to modify shell configuration files (
~/.bashrcor~/.zshrc) to add aliases for thekubectl-mcp-servertool. While standard for CLI tools, this involves the modification of persistent environment files. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from a Kubernetes cluster that may be influenced by external actors.
- Ingestion points: Data enters the agent context through tools like
kubectl_get,kubectl_logs, and MCP resources such aspods://{namespace},events://{namespace}, andservices://{namespace}. - Boundary markers: There are no explicit instructions or delimiters defined to separate cluster data from agent instructions.
- Capability inventory: The skill includes high-privilege capabilities such as scaling deployments (
kubectl_scale) and installing software via Helm (install_helm_chart). - Sanitization: No sanitization or validation of the ingested cluster data is specified in the skill instructions beyond an environment variable for masking secrets (
MCP_MASK_SECRETS).
Audit Metadata