openai-realtime-voice

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a security-focused design framework for voice-based AI applications, prioritizing user privacy, safety boundaries, and least-privilege tool access.
  • [EXTERNAL_DOWNLOADS]: The skill references official OpenAI documentation and API endpoints (api.openai.com) for legitimate integration and authorization purposes.
  • [REMOTE_CODE_EXECUTION]: The provided code examples utilize established and verified Node.js packages, including express, openai, ws, and @openai/agents, which are standard for building these types of applications.
  • [PROMPT_INJECTION]: The skill instructs developers to implement defensive prompts that manage user-provided audio securely, including explicit confirmation steps for side-effecting tools and clear boundaries for sensitive information.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 01:52 PM
Security Audit — agent-trust-hub — openai-realtime-voice