nano-banana-pro

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes official and well-known libraries google-genai and pillow for API interaction and image manipulation.
  • [COMMAND_EXECUTION]: It executes a bundled Python script generate_image.py using the uv package manager to perform its tasks.
  • [DATA_EXFILTRATION]: The skill transmits user-specified prompt text and local images to Google's well-known Gemini API; this is intended behavior and uses trusted infrastructure.
  • [PROMPT_INJECTION]: The tool is subject to indirect prompt injection as it processes external inputs (Ingestion points: --prompt and -i arguments in SKILL.md; Boundary markers: None; Capability inventory: generate_image.py performs network API calls and file system read/write; Sanitization: None).
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:31 PM
Security Audit — agent-trust-hub — nano-banana-pro