doca-dpa-hl-tracer
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a documentation-only artifact authored by NVIDIA, referencing official NVIDIA documentation and repositories. No malicious patterns such as obfuscation, hardcoded credentials, or unauthorized network operations were detected.
- [COMMAND_EXECUTION]: The skill provides structured workflows for executing the
doca_dpa_hl_tracerhost-side CLI. It correctly routes the agent to official documentation for handling privilege requirements and mandatory device arguments, ensuring safe tool interaction. - [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection (Category 8) due to its core diagnostic requirements.
- Ingestion points: The agent is instructed to read and analyze human-readable log files and parsed event outputs generated from DPA-side trace captures (as detailed in
TASKS.mdandCAPABILITIES.md). - Boundary markers: The instructions do not prescribe the use of delimiters or specific safety warnings to ignore embedded instructions within the ingested trace data.
- Capability inventory: The skill facilitates the execution of the tracer CLI for data capture and parsing but does not involve arbitrary script generation.
- Sanitization: No specific sanitization or filtering logic is provided for the trace content before agent analysis; the agent is instead encouraged to quote captured events verbatim.
Audit Metadata