mrbeast-perspective

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides utility scripts in the scripts/ directory for analyzing video content. These scripts perform standard file reading and pattern matching operations using Python and Shell commands to audit content metadata.
  • [EXTERNAL_DOWNLOADS]: The fetch_youtube_subtitles.sh script downloads subtitles from YouTube using the well-known yt-dlp utility. It also includes an automated installation routine for yt-dlp from the official PyPI registry if the tool is not found locally. These operations are transparent and target well-known services.
  • [SAFE]: The persona configuration in SKILL.md establishes a role-play environment for content advice. There are no instructions that attempt to override safety filters, extract system prompts, or manipulate the agent's core behavior beyond the stated persona.
  • [SAFE]: Analysis of the provided scripts shows that they process user-supplied text and URLs strictly for the intended purpose of content evaluation. No sensitive file access (e.g., credentials or SSH keys) or unauthorized data exfiltration patterns were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 03:23 AM