youtube-automation

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Prompt Injection (LOW): Potential for indirect prompt injection via untrusted data ingestion. Ingestion points: YOUTUBE_SEARCH_YOU_TUBE, YOUTUBE_LIST_COMMENT_THREADS, and YOUTUBE_VIDEO_DETAILS. Boundary markers: Absent; no instructions provided to ignore embedded commands within retrieved content. Capability inventory: Includes impactful actions such as YOUTUBE_UPLOAD_VIDEO, YOUTUBE_UPDATE_VIDEO, and YOUTUBE_SUBSCRIBE_CHANNEL. Sanitization: Absent; external data is likely processed directly by the agent without filtering.
  • External Downloads (SAFE): The skill utilizes https://rube.app/mcp as its MCP server. While the domain is not in the trusted list, it is a configuration endpoint necessary for the skill's defined functionality and does not involve direct local script execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 03:10 PM