veo-hardened

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script (generate_video.py) via the uv package manager to process video generation requests.
  • [EXTERNAL_DOWNLOADS]: The Python script downloads generated video files from Google's official API servers using the google-genai library.
  • [SAFE]: The skill instructions direct the agent to retrieve an API key from a specific local configuration file while providing guardrails to limit access to only the necessary credential field.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 06:32 PM
Security Audit — agent-trust-hub — veo-hardened