k8s-launch-kit-deploy
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
l8kcommand-line tool to deploy networking manifests and useskubectlto verify resource status within the cluster.\n- [CREDENTIALS_UNSAFE]: The skill accesses the cluster configuration at~/.kube/configand through theKUBECONFIGenvironment variable to perform administrative actions, which is required for its primary function.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates the download and installation of thenvidia/network-operatorHelm chart from the vendor's repository as part of the deployment process.\n- [PROMPT_INJECTION]: The skill ingests YAML manifests from user-defined local directories, creating an indirect prompt injection attack surface.\n - Ingestion points: Manifest files located in the directory specified by the
--deployment-filesflag inSKILL.md.\n - Boundary markers: No specific delimiters or warnings to ignore embedded instructions are provided.\n
- Capability inventory: The skill possesses cluster-write capabilities through the
l8k deploycommand and read access viakubectl.\n - Sanitization: No explicit sanitization or validation of the content of the manifest files is described.
Audit Metadata