youtube
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: All network operations are directed to official, trusted Google API domains (https://www.googleapis.com), minimizing risks of data exfiltration.
- [SAFE]: Authentication is managed securely through a platform-provided OAuth token in an environment variable ($GOOGLE_YOUTUBE_TOKEN), avoiding hardcoded secrets.
- [SAFE]: The skill includes clear instructions for the agent to seek user confirmation before performing sensitive actions like publishing new videos.
- [SAFE]: The skill processes untrusted data from YouTube (comments and video metadata), which represents a managed indirect prompt injection surface. Ingestion points: Video search results and comment threads via API. Boundary markers: The agent uses structured JSON parsing. Capability inventory: Limited to Bash with curl and jq. Sanitization: Standard JSON handling via jq is used for output processing.
Audit Metadata