gemini-live-api-dev

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs users to install the official google-genai (Python) and @google/genai (Node.js) SDKs from standard registries. It also references technical documentation hosted on Google's official developer portal (ai.google.dev). These sources are recognized as trusted.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents an interface for processing real-time, user-provided data streams. This attack surface is mitigated by explicit instructions in the 'Anti-Patterns' section of SKILL.md, which directs the agent not to treat external content or tool outputs as trusted instructions.
  • [CREDENTIALS_UNSAFE]: Code examples in SKILL.md use generic placeholders (YOUR_API_KEY) for authentication parameters, following safe documentation practices. The skill further promotes security by recommending the use of ephemeral tokens for client-side applications.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:52 AM
Security Audit — agent-trust-hub — gemini-live-api-dev