zoom-rtms

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFE
Full Analysis
  • [Standard Networking]: The skill documents connections to official Zoom WebSocket infrastructure and established third-party AI services such as OpenAI, Deepgram, and AssemblyAI. These connections are necessary for the skill's stated purpose of real-time media ingestion and analysis.
  • [Credential Management]: Implementation examples consistently use environment variables for managing API keys and secrets, aligning with security best practices for protecting sensitive credentials.
  • [Data Ingestion Surface]: The skill is designed to process real-time meeting data, including audio and transcripts. While this represents a surface for indirect prompt injection, the skill does not implement high-risk capabilities like arbitrary command execution or system-level modifications, resulting in a safe profile for its intended use case.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 11:37 PM