nano-banana-pro-product-image

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions and a Python script (scripts/imagegen.py) intended to be executed via the command line to perform image generation tasks, including initialization, image generation, and task status polling.
  • [EXTERNAL_DOWNLOADS]: The scripts/imagegen.py script contains functionality to download generated images and media results from the AI Hive service (ai-hive.iclip.cn) to the local filesystem (defaulting to ~/Downloads/AiHive).
  • [DATA_EXPOSURE_&_EXFILTRATION]: The skill manages an API key (AI_HIVE_API_KEY) for authentication. It implements a secure local storage mechanism in ~/.ai-hive/config.json using chmod 600 to prevent unauthorized access by other users on the system.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided text prompts and external image references which are interpolated into API requests. While this presents a surface for injection, the skill acts as a pass-through to the backend model service without performing high-privilege local operations based on the model's output.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 12:20 PM
Security Audit — agent-trust-hub — nano-banana-pro-product-image