social-media-scout

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill manages a TikHub API key which is stored locally in scripts/config.json. The tikhub_client.py script transmits this key in the Authorization header to api.tikhub.io to authenticate requests. This is the intended behavior for the tool's primary function and targets the official API service.
  • [COMMAND_EXECUTION]: Instructions in SKILL.md guide the agent to install and run yt-dlp, an established third-party utility, for downloading videos from Bilibili. This interaction is transparently documented and occurs as part of the tool's video processing workflow.
  • [EXTERNAL_DOWNLOADS]: The tikhub_client.py script includes a download_file function that facilitates downloading social media video content to the local filesystem using standard Python libraries. This function is intended for processing media files as requested by the user.
  • [PROMPT_INJECTION]: The skill ingests untrusted content from social media platforms (e.g., video descriptions, comments). While this creates a surface for indirect prompt injection, the skill acts as a data retriever and does not execute the retrieved content, making the risk inherent to the processing of public social media data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 05:56 AM