Skills
Navigate
skills/maystudios/claude-skills/video-fetch-and-summarize/Gen Agent Trust Hub

video-fetch-and-summarize

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DYNAMIC_EXECUTION]: The script uses __import__ to verify the presence of required Python modules like yt-dlp and google-genai at runtime.
  • [COMMAND_EXECUTION]: The skill executes pip for dependency installation and yt-dlp for video processing using subprocess.run with list-based arguments.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted video content from external URLs and interpolates user-supplied prompts into LLM requests.
  • Ingestion points: External video URLs and the --prompt argument in scripts/fetch_and_summarize.py.
  • Boundary markers: No explicit delimiters are used to wrap user-provided prompts.
  • Capability inventory: Subprocess execution, network access, and file system writes.
  • Sanitization: Folder names are sanitized via regex; URL extraction is performed using regex validation.
  • [EXTERNAL_DOWNLOADS]: The skill automatically installs necessary Python packages (yt-dlp, google-genai) from the standard package registry if they are missing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 05:53 AM