seedream-5-lite-character-consistency

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements secure secret management for API keys. The init command in scripts/imagegen.py creates a configuration file at ~/.ai-hive/config.json and explicitly restricts its permissions to 0600 (read/write by owner only) using os.chmod, preventing other users on the system from accessing the credentials.- [EXTERNAL_DOWNLOADS]: The script performs network requests to the AI Hive API (ai-hive.iclip.cn) to upload reference media and download generated image or video results. These operations are the primary purpose of the skill and utilize standard HTTP protocols via the requests library.- [COMMAND_EXECUTION]: The init command uses the Python webbrowser module to open the official API access page. This is a benign interactive feature designed to guide the user through the onboarding process.- [DATA_EXPOSURE]: The skill accesses the local file system to read reference images provided via the --image flag. This behavior is transparent and controlled by the parameters passed to the script.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 12:20 PM
Security Audit — agent-trust-hub — seedream-5-lite-character-consistency