nano-banana-pro

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill automatically manages dependencies such as google-genai and pillow through the uv tool, fetching them from the official Python Package Index (PyPI).
  • [COMMAND_EXECUTION]: The skill provides commands for the agent to execute the scripts/generate_image.py script locally to process image generation and editing requests.
  • [DATA_EXFILTRATION]: The script sends user-provided prompts and image data to Google's official Gemini API. This communication is required for the skill's core functionality and involves a well-known service provider.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external prompts and image files.
  • Ingestion points: The script scripts/generate_image.py reads user input via the --prompt argument and image files via the --input-image argument.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the script logic.
  • Capability inventory: The skill has the capability to read local files, write generated images to the disk using PIL, and make network requests to the Gemini API.
  • Sanitization: No sanitization or validation is performed on the prompt strings or image metadata before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 09:05 AM