douyin-downloader
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by scraping HTML from Douyin share links.\n
- Ingestion points: External HTML content from iesdouyin.com is fetched and parsed in scripts/parse_douyin.py.\n
- Boundary markers: The instructions do not define specific delimiters for separating untrusted content from agent instructions.\n
- Capability inventory: The skill has the capability to write files to the local file system through scripts/download_video.py and upload file contents to a remote API in scripts/transcribe_audio.py.\n
- Sanitization: Filenames are sanitized in scripts/parse_douyin.py using a sanitize_title function which removes potentially dangerous characters.\n- [EXTERNAL_DOWNLOADS]: The skill performs network requests to Douyin (iesdouyin.com, douyin.com, v.douyin.com) to extract video metadata and download content. It also communicates with Volcano Engine (openspeech.bytedance.com) for speech-to-text transcription. These are established services consistent with the skill's functionality.\n- [COMMAND_EXECUTION]: The skill utilizes Python scripts (parse_douyin.py, download_video.py, transcribe_audio.py) to handle media processing. These scripts are invoked via standard command-line interfaces and do not involve arbitrary shell command execution.
Audit Metadata