gpt-image-2-ad-image
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions as intended for its documented purpose without exhibiting malicious behaviors. All network and file operations are transparent and necessary for the core functionality of the tool.\n- [EXTERNAL_DOWNLOADS]: The skill requires the
requestslibrary, a standard and widely-trusted Python package for handling HTTP requests.\n- [COMMAND_EXECUTION]: The skill executes a local Python script (scripts/imagegen.py) to handle API logic, media uploads, and result downloads, which is the standard operation mode for this tool.\n- [DATA_EXFILTRATION]: Prompts and reference images are sent to the official service domainai-hive.iclip.cn. This activity is central to the skill's purpose and does not target unauthorized or suspicious endpoints.\n- [CREDENTIALS_UNSAFE]: The skill provides a secure initialization command (init) that stores API keys in~/.ai-hive/config.json. The script correctly sets file permissions to0600, ensuring the configuration is restricted to the current user.
Audit Metadata