fishjam-js-server-sdk

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the @fishjam-cloud/js-server-sdk library and the optional @google/genai dependency. These are official packages from the vendor and a well-known service provider respectively.
  • [SAFE]: Promotes secure credential handling by utilizing environment variables for management tokens and providing explicit warnings against exposing them in client-facing code.
  • [SAFE]: Implements secure webhook handling patterns through path-based secrets, ensuring that only authorized notifications from the Fishjam platform are processed.
  • [SAFE]: Provides a structured surface for AI agent development with well-defined data ingestion and capability boundaries.
  • [SAFE]: (Indirect Prompt Injection Surface Analysis) The skill enables processing of external media data. Ingestion points: Receives audio via agent.on('trackData') and video frames via agent.captureImage() in agent.md. Boundary markers: Relies on the integrated AI model's internal safety filters as no explicit prompt delimiters are provided for the audio/image streams. Capability inventory: Includes capabilities to manipulate room state and send media data via the FishjamClient and FishjamAgent classes. Sanitization: Standard for media SDKs, it passes raw binary data to processors without content-level sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 01:40 PM
Security Audit — agent-trust-hub — fishjam-js-server-sdk