video-gen

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements video generation as its primary function using established API patterns. It handles user inputs (prompts and images) to generate and download video files to the local project directory.
  • [EXTERNAL_DOWNLOADS]: The script downloads the generated video files from the FAL API infrastructure (fal.media) to the user's project directory. This is a core functionality of the skill and uses the designated API provider infrastructure.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user prompts and image files for processing.
  • Ingestion points: Input is provided via the --prompt, --start-image, and --end-image CLI arguments in the video-gen.py script.
  • Boundary markers: The instructions in SKILL.md mandate the use of AskUserQuestion to gather specific parameters, ensuring the agent operates within user-defined bounds.
  • Capability inventory: The skill uses network access for API communication, reads local files for uploads, and writes files for video downloads.
  • Sanitization: Inputs are passed as standard data fields in a JSON payload to the API proxy, preventing them from being interpreted as instructions by the skill's own execution environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 01:35 PM