The Agent Skills Directory

[SAFE]: The skill follows secure patterns for environment inspection and setup delegation. No malicious intent, obfuscation, or unauthorized access patterns were detected.
[COMMAND_EXECUTION]: The skill executes standard diagnostic commands (nvidia-smi, uname, docker --version, dpkg -l) to assess host compatibility. This behavior is consistent with the stated purpose of guiding an SDK installation.
[EXTERNAL_DOWNLOADS]: Fetches platform compatibility and release metadata from docs.nvidia.com. This is a trusted documentation source used to ensure the agent uses up-to-date installation requirements.
[DATA_EXFILTRATION]: The helper script scripts/check_conda.sh reads shell configuration files (~/.bashrc, ~/.zshrc, etc.) to locate Conda installation paths. This is a local read operation used for environment detection; no contents are exfiltrated or sent to external endpoints.

holoscan-setup