happy-horse-reference-to-video
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a Python CLI script to interact with the AI Hive OpenAPI. The script performs standard operations including model lookup, task submission, and progress polling at the vendor's domain, ai-hive.iclip.cn.
- [DATA_EXFILTRATION]: The script uploads user-selected media files to the AI Hive API for processing. This behavior is the core function of the skill and is documented for the user.
- [EXTERNAL_DOWNLOADS]: Upon completion of a generation task, the skill downloads result files from the service's result URLs to the local filesystem (specifically the ~/Downloads/AiHive directory).
- [CREDENTIALS_UNSAFE]: The skill implements secure local storage for the AI Hive API key. It creates a configuration file at ~/.ai-hive/config.json and programmatically enforces a 0600 permission mask to ensure the key is only accessible by the current user.
- [COMMAND_EXECUTION]: The skill instructions guide the user to execute the videogen.py script. The script's logic is confined to handling API communications and file operations related to media generation, with no evidence of arbitrary command execution or malicious system modifications.
Audit Metadata