skills/nvidia/skills/doca-rdma/Gen Agent Trust Hub

doca-rdma

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes command-line tools for environment discovery and diagnostic purposes. Evidence: Use of pkg-config, doca_caps, dmesg, mlxconfig, and ibv_devinfo in TASKS.md for configuring and debugging the RDMA environment.
  • [EXTERNAL_DOWNLOADS]: The skill references external resources for documentation and code samples. Evidence: Mentions of docs.nvidia.com and github.com/NVIDIA-DOCA/doca-samples are directed at official vendor repositories and documentation sites, supporting the legitimate development workflow.
  • [SAFE]: Diagnostic commands requiring elevated privileges (sudo) are documented for hardware-level monitoring and configuration, which is standard for DPU/NIC administration and consistent with the skill's primary purpose.
  • [SAFE]: Indirect Prompt Injection analysis indicates standard data processing surfaces. Ingestion points: User project context and shell command outputs (TASKS.md). Boundary markers: Instructions emphasize starting from verified local samples (SKILL.md). Capability inventory: Shell command execution for diagnostics (TASKS.md). Sanitization: Guidance relies on standard library usage and verified sample patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 05:55 AM
Security Audit — agent-trust-hub — doca-rdma