youtube-automation
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it ingests untrusted data from external sources and possesses capabilities to modify account state.\n
- Ingestion points: Untrusted data enters the agent context via
YOUTUBE_LIST_COMMENT_THREADSandYOUTUBE_SEARCH_YOU_TUBEas described inSKILL.md.\n - Boundary markers: The instructions do not define clear delimiters or include warnings to ignore instructions embedded within the retrieved YouTube content.\n
- Capability inventory: The skill provides access to impactful tools including
YOUTUBE_UPLOAD_VIDEO,YOUTUBE_UPDATE_VIDEO,YOUTUBE_CREATE_PLAYLIST, andYOUTUBE_SUBSCRIBE_CHANNEL(referenced inSKILL.md).\n - Sanitization: There is no evidence of sanitization or validation of the ingested external content before it is processed by the agent.\n
- Mitigation: Sanitize external content before processing, use explicit boundary markers to separate untrusted data, and require human review for write operations triggered by external data.\n- [EXTERNAL_DOWNLOADS]: The skill requires connection to an external MCP server endpoint at
https://rube.app/mcp. This is a remote dependency required for core functionality.
Audit Metadata