video-style-analysis

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: Multiple scripts in the pipeline (asr_transcriber.py, audio_video_mixer.py, scene_concat.py, video_generator.py) use subprocess.run to execute FFmpeg for audio extraction, video stitching, and resizing. These operations are required for the skill's media processing functionality.
  • [COMMAND_EXECUTION]: The orchestration script perfect_replication_workflow.py uses subprocess.run to manage the execution of other Python scripts within the local project directory.
  • [EXTERNAL_DOWNLOADS]: The api_client.py utility downloads generated image and video content from the VolcEngine and ApiMart APIs. VolcEngine is a recognized trusted organization, and these downloads are the intended output of the generation process.
  • [EXTERNAL_DOWNLOADS]: The asr_transcriber.py script downloads Whisper models for speech-to-text processing using the official openai-whisper library interface.
  • [CREDENTIALS_UNSAFE]: The skill manages API keys securely by using placeholders in example files (api_keys.yaml.example) and instructing users to provide actual keys through environment variables or local, non-versioned configuration files.
  • [DATA_EXFILTRATION]: The skill transmits metadata, prompts, and reference images to external generation APIs. This data transfer is fundamental to the skill's core functionality of recreating video styles and does not target sensitive user data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 02:54 PM
Security Audit — agent-trust-hub — video-style-analysis