video-processing-editing

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses list-based subprocess execution for FFmpeg and FFprobe commands across all scripts (e.g., video_editor.py, audio_mixer.py), which effectively mitigates shell injection risks by avoiding shell=True.
  • [SAFE]: No hardcoded credentials, sensitive file paths, or suspicious network operations were detected. All network-related commands found in the documentation (like curl) are examples of legitimate configuration or resource fetching from standard sources.
  • [SAFE]: The skill contains no obfuscated code, persistence mechanisms, or attempts at privilege escalation. It strictly focuses on video processing automation as described in its metadata.
  • [SAFE]: Although the skill processes external video and audio data, it does not interpolate this data into LLM prompts, significantly reducing the risk of indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 02:44 AM