douyin-video-fetch

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill runs a localized Python script to facilitate browser automation and file management.
  • [EXTERNAL_DOWNLOADS]: Retrieves video data and JSON metadata from Douyin's official domains. It utilizes the playwright library to manage browser-based interactions.
  • [DATA_EXFILTRATION]: Saves video files to local storage. While the output path is configurable via the --output argument, the script primarily targets the temporary directory (/tmp).
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes external content from the internet:
  • Ingestion points: fetch_detail in scripts/fetch_douyin_video.py ingests data from Douyin's web pages and API endpoints.
  • Boundary markers: None identified; untrusted content is not delimited.
  • Capability inventory: The skill includes file writing (download_video) and network request (urllib, playwright) capabilities.
  • Sanitization: The script performs no sanitization on the metadata (e.g., video descriptions) extracted from external sources before outputting it.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 06:59 PM