NYC

bilibili-downloader

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE] (HIGH): The skill documentation mentions a --cookies parameter for 'getting high-definition quality'. Accessing browser cookies is a high-risk operation that exposes sensitive session data and authentication tokens.
  • [COMMAND_EXECUTION] (MEDIUM): The skill functions by generating and executing shell commands (e.g., python ~/.claude/skills/bilibili-downloader/bili_download.py --url "..."). If the input url or space parameters are not strictly sanitized, this provides a surface for command injection.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill requires the installation of yt-dlp via pip and downloads media content from external servers (Bilibili). While Bilibili is a known site, the automated download of external binary data poses a risk.
  • [DATA_EXPOSURE] (LOW): The skill defaults to saving files to the user's desktop (~/Desktop/bilibili_downloads), which involves file system writes and potential exposure of directory structures.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:37 PM