doca-debug
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured technical documentation for debugging the NVIDIA DOCA SDK. It prioritizes non-destructive, read-only diagnostic steps and correctly identifies proper escalation paths to official NVIDIA forums. The instructions include security-conscious guidance, such as redacting sensitive payload data from core dumps before posting to public forums.
- [EXTERNAL_DOWNLOADS]: The skill references official NVIDIA resources including the DOCA SDK documentation, official GitHub repositories (NVIDIA-DOCA/doca-samples, NVIDIA/doca-platform), and the NVIDIA Developer Forum. These are established vendor resources and well-known service domains.
- [PROMPT_INJECTION]: The skill facilitates the ingestion and analysis of untrusted diagnostic data (application logs and system output), representing an indirect prompt injection surface. This behavior is documented neutrally as it is central to the skill's primary purpose.
- Ingestion points: Workflow steps in
TASKS.mdinvolve processing user-supplied program output (stderr), system logs (dmesg), and service journals (journalctl). - Boundary markers: The instructions do not define specific delimiters for separating processed log content from the agent's internal reasoning context.
- Capability inventory: The agent uses results of log analysis to recommend diagnostic commands or escalate to support forums.
- Sanitization: No explicit sanitization of log contents is specified within the instructions.
Audit Metadata