image-2-text-to-image

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a legitimate interface for the AI Hive image generation service. The code is well-structured, follows security best practices for configuration management, and contains no malicious patterns.
  • [EXTERNAL_DOWNLOADS]: The script downloads generated media assets (images and videos) from the official service domain ai-hive.iclip.cn. These downloads are the intended results of the skill's operation.
  • [COMMAND_EXECUTION]: The skill requires the standard requests library and provides a Python script to interact with the API. The instructions involve running the script via the command line, but the script itself does not perform any unauthorized or dangerous system operations.
  • [CREDENTIALS_UNSAFE]: API keys are managed safely. The script stores the user-provided key in ~/.ai-hive/config.json and explicitly sets file permissions to 0600 (read/write by owner only) to prevent unauthorized access by other local users.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 12:20 PM
Security Audit — agent-trust-hub — image-2-text-to-image