ai-video-production-master

Warn

Audited by Gen Agent Trust Hub on Nov 2, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/cloud_i2v_batch.py uses the subprocess module to execute system commands including the vastai CLI for instance management and ssh/scp for remote file operations and instance communication.
  • [COMMAND_EXECUTION]: The script scripts/motion_graphics_generator.py invokes ffmpeg, rsvg-convert, and ImageMagick's convert via subprocess.run to generate and animate video elements from SVG templates.
  • [CREDENTIALS_UNSAFE]: The scripts/cloud_i2v_batch.py script programmatically searches for and accesses private SSH keys in the user's home directory (specifically ~/.ssh/id_ed25519_vastai, ~/.ssh/id_ed25519, and ~/.ssh/id_rsa). These keys are used to authenticate and facilitate file transfers with cloud-based GPU instances.
  • [EXTERNAL_DOWNLOADS]: The skill's cloud orchestration script (scripts/cloud_i2v_batch.py) configures remote instances to download approximately 18GB of AI models and encoders from Hugging Face's official repositories using wget during the setup phase.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a broad attack surface for indirect prompt injection as it ingests untrusted data from user-provided script files and image keyframes. This data is interpolated into prompts for generative AI models. The skill lacks explicit boundary markers or sanitization logic to mitigate the risk of malicious instructions embedded in input data influencing the agent's behavior or subsequent command execution steps.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 2, 2026, 09:42 AM