dvar-torah-animation

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads AI-generated images and video segments from the official Google Generative Language API (Gemini/Veo). These requests are directed to a trusted service provider.\n- [COMMAND_EXECUTION]: Local scripts execute ffmpeg and ffprobe binaries via subprocess.run to perform media processing tasks such as scaling, crossfading, and audio time-stretching. These operations are core to the skill's primary functionality and are performed on locally generated or provided assets.\n- [DYNAMIC_EXECUTION]: The skill dynamically constructs command-line arguments and complex filter chains for ffmpeg at runtime, enabling automated video editing based on calculated timing anchors.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text which is subsequently used to generate prompts for generative AI models. While this represents a potential surface for indirect prompt injection, the risk is minimal as the content only influences the visual and audio assets generated by isolated external models.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 06:39 PM