detecting-qr-code-phishing-with-email-security

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a functional toolset for cybersecurity analysis, specifically targeting QR code phishing detection in emails and images.
  • [SAFE]: Python scripts (agent.py, process.py) utilize well-known libraries such as Pillow and pyzbar for image processing and QR decoding. No suspicious command execution or dynamic code evaluation (eval/exec) patterns were found.
  • [SAFE]: Data ingestion is limited to reading local .eml and image files for analysis. The tool does not perform unauthorized network requests or exfiltrate sensitive data.
  • [SAFE]: API references in documentation use standard placeholders for credentials, and no hardcoded secrets were found in the source code.
  • [SAFE]: While the skill processes untrusted input (emails/images), it does so via static regex analysis and library-based decoding rather than feeding raw content into a prompt for LLM execution, minimizing indirect prompt injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 06:46 PM