nano-banana-pro

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements image generation and editing functionality using the official google-genai library and the gemini-3-pro-image-preview model.
  • [SAFE]: External dependencies include uv, google-genai, and pillow, which are well-known and standard tools in the Python ecosystem.
  • [SAFE]: Sensitive data such as the GEMINI_API_KEY is handled securely via environment variables or local configuration files, with instructions pointing to standard paths like ~/.openclaw/openclaw.json.
  • [SAFE]: File system access is limited to reading user-specified input images and writing the generated output image to a path provided in the command-line arguments.
  • [SAFE]: Network operations are restricted to communication with the Google Generative AI API for its intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 12:14 AM