skills/eat-pray-ai/yutu/youtube/Gen Agent Trust Hub

youtube

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install the yutu CLI from vendor-managed sources including the @eat-pray-ai/yutu NPM package and the github.com/eat-pray-ai/yutu repository.
  • [COMMAND_EXECUTION]: Uses the yutu command to interact with the YouTube Data API for operations ranging from video uploads to content deletion.
  • [DATA_EXFILTRATION]: Accesses sensitive authentication files such as client_secret.json and youtube.token.json, which are used to store OAuth credentials and tokens.
  • [PROMPT_INJECTION]: Vulnerable to indirect prompt injection due to the handling of external data from YouTube.
  • Ingestion points: Data entering the context from yutu search, comment, video, and activity list commands.
  • Boundary markers: No delimiters or ignore instructions are present for retrieved content.
  • Capability inventory: Destructive operations like yutu video delete and yutu comment delete are available.
  • Sanitization: No specific sanitization or content validation procedures are documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 10:17 PM