gemini-live-api-dev

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • External Documentation and Resources: The skill references official documentation and lists several established technology partners for real-time integrations. These references provide necessary context for API development and trace back to official infrastructure.
  • Data Ingestion Surface: The skill provides instructions for processing real-time user input, including text, audio, and video streams. 1. Ingestion points: User data is transmitted through the send_realtime_input method in SKILL.md. 2. Boundary markers: There are no specific delimiters mentioned for the streamed media content. 3. Capability inventory: The skill utilizes WebSocket connections for low-latency, bidirectional communication. 4. Sanitization: Standard for such APIs, the documentation focuses on session handling rather than content filtering at the transmission layer.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:23 AM
Security Audit — agent-trust-hub — gemini-live-api-dev