slurm

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues, malicious code, or suspicious patterns were detected in the skill file. The content consists of benign documentation and operational guidelines for SLURM usage.- [PROMPT_INJECTION]: The skill provides instructional content for agent behavior without any attempts to override safety filters, extract system prompts, or implement jailbreak-style instructions.- [COMMAND_EXECUTION]: While the skill references SLURM command-line tools like sbatch and sacct, it explicitly mandates that the agent must never submit jobs without explicit user confirmation, adhering to safety best practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 08:19 PM
Security Audit — agent-trust-hub — slurm