dgx-station-diagnose

Pass

Audited by Gen Agent Trust Hub on Oct 3, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a shell script wrapper (scripts/dgx-assist) to execute a local diagnostic binary. It passes arguments like report IDs and finding IDs to perform targeted analysis and fixes.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests structured JSON output from local diagnostic runs to interpret hardware and software health. This injection surface is mitigated by limiting modifications to a specific allowlisted registry and requiring explicit human approval with a dry-run preview before any state-changing command is executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 3, 2026, 05:26 PM
Security Audit — agent-trust-hub — dgx-station-diagnose