social-media-publishing

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes ffmpeg via subprocess.run to encode videos for native Reddit uploads. This is found in scripts/reddit/native_video.py and is a legitimate part of the video processing workflow.
  • [COMMAND_EXECUTION]: The YouTube CLI uses os.execve in scripts/youtube/cli.py to re-execute the script using a specific Python virtual environment to ensure the modal library is available for uploads.
  • [EXTERNAL_DOWNLOADS]: Several platform scripts (LinkedIn, X, TikTok) support downloading media assets from user-supplied public HTTPS URLs. This functionality uses urllib.request.urlopen to fetch assets before uploading them to the respective platforms.
  • [SAFE]: The skill implements best practices for secret management by loading credentials from environment variables or dedicated configuration files in the user's home directory (e.g., ~/.secrets/reddit/env). No hardcoded secrets or unsafe credential handling were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 11:06 AM
Security Audit — agent-trust-hub — social-media-publishing