research-grants

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • PROMPT_INJECTION (SAFE): No override markers or jailbreak patterns detected. Instructions are purely for grant writing formatting.
  • EXTERNAL_DOWNLOADS (SAFE): External links refer only to official government research portals (.gov). No automated downloads or script executions are initiated.
  • DATA_EXFILTRATION (SAFE): No hardcoded secrets, sensitive path access, or exfiltration patterns identified.
  • REMOTE_CODE_EXECUTION (SAFE): No package installations or script execution commands found.
  • NO_CODE (SAFE): The skill is comprised of text-based reference materials and project templates, posing no functional security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:43 PM