Skills
skills/honeybeecnc/youtube-clipper-skill/youtube-clipper/Gen Agent Trust Hub

youtube-clipper

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Prompt Injection (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it processes untrusted subtitle data from YouTube.\n
  • Ingestion points: Subtitles are downloaded in scripts/download_video.py.\n
  • Boundary markers: Weak delimiters are used in scripts/translate_subtitles.py and scripts/generate_summary.py when presenting data to the AI.\n
  • Capability inventory: The skill has file system access and can execute FFmpeg via subprocess.run.\n
  • Sanitization: Filenames are sanitized in scripts/utils.py, and subtitle text is cleaned of HTML tags in scripts/analyze_subtitles.py.\n- External Downloads (LOW): The install_as_skill.sh script installs yt-dlp, pysrt, and python-dotenv from PyPI. These are standard packages required for the skill's operation.\n- Command Execution (LOW): The skill uses subprocess.run with list arguments to call ffmpeg and yt-dlp. This avoids shell injection vulnerabilities but maintains dependency on external binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:21 PM