troubleshoot-nvidia-dcgm
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows security best practices for diagnostic agents. It uses read-only queries to fetch telemetry data through an established MCP interface.
- [DATA_EXFILTRATION]: The skill accesses system-level telemetry (GPU health, thermal data, errors) via Netdata MCP tools. Analysis shows no evidence of data exfiltration to external domains; queries are scoped to diagnostic signals.
- [PROMPT_INJECTION]: The instructions do not contain patterns for bypassing safety guidelines or overriding agent behavior. The logic is strictly limited to the stated purpose of NVIDIA DCGM troubleshooting.
Audit Metadata