youtube-automation

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Connects to an external MCP server at https://rube.app/mcp to retrieve tool schemas and definitions for YouTube automation.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through the processing of untrusted external content combined with sensitive write capabilities.
  • Ingestion points: Processes untrusted data from YouTube via YOUTUBE_LIST_COMMENT_THREADS and YOUTUBE_SEARCH_YOU_TUBE (SKILL.md).
  • Boundary markers: The skill does not define delimiters or instructions for the agent to ignore potential commands embedded in external content.
  • Capability inventory: Includes sensitive account-level operations such as YOUTUBE_UPLOAD_VIDEO, YOUTUBE_UPDATE_VIDEO, and YOUTUBE_SUBSCRIBE_CHANNEL (SKILL.md).
  • Sanitization: No sanitization or validation of the content retrieved from YouTube is described prior to processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 07:01 AM