regulatory-affairs-head

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's instructions and reference files are purely informational and procedural, focusing on medical device compliance. No malicious instructions or hidden patterns were detected.
  • [DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file access, or unauthorized data exfiltration patterns was found. All external references are to official regulatory bodies and standard databases.
  • [COMMAND_EXECUTION]: The skill references a local Python utility for submission tracking, but there are no instructions to execute arbitrary or untrusted code from external sources. The script mention is for project management purposes.
  • [PROMPT_INJECTION]: No attempts to bypass AI safety guidelines, override system behavior, or extract instructions were identified.
  • [EXTERNAL_DOWNLOADS]: No remote scripts, external packages, or unauthorized downloads were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 08:06 PM
Security Audit — agent-trust-hub — regulatory-affairs-head