doca-telemetry-utils
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely documentation-based and does not contain any executable scripts or binary files.
- [SAFE]: No external network operations or remote code downloads were detected. References to
docs.nvidia.comare for documentation purposes only. - [SAFE]: The command patterns provided (
doca_telemetry_utils get-counters, etc.) are standard diagnostic tools for the DOCA environment and operate in a read-only manner according to the skill's safety policy. - [SAFE]: No sensitive data exposure, persistence mechanisms, or privilege escalation attempts were found. The skill correctly instructs users to follow host policies for necessary elevated privileges when probing devices.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied counter names and property values into command-line arguments. However, it includes strong boundary markers, instructing the agent to 'quote only real symbols/commands' and 'never construct Data IDs by hand,' effectively mitigating injection risks.
Audit Metadata