skills/nvidia/openshell/openshell-cli/Gen Agent Trust Hub

openshell-cli

Pass

Audited by Gen Agent Trust Hub on Apr 27, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents commands that execute code within sandboxed environments and the management gateway. For example, openshell sandbox create -- <command> executes commands in a new sandbox, and openshell doctor exec -- <command> allows running commands inside the gateway container.
  • [CREDENTIALS_SAFE]: The provider command group manages API keys and tokens. The documentation promotes secure practices by recommending the --from-existing flag to load credentials from local state or using environment variables instead of hardcoded strings.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the use of external container images and Dockerfiles through the openshell sandbox create --from <source> command. This is a standard feature for 'Bring Your Own Container' (BYOC) workflows.
  • [DATA_EXPOSURE]: The skill mentions accessing SSH keys (e.g., ~/.ssh/id_rsa) for remote cluster deployment. This is a standard and necessary operation for the described administrative tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 27, 2026, 08:08 AM