zoom-plugin-sdk-macos

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents an interface for processing untrusted external data within Zoom meetings, creating a potential surface for indirect prompt injection if the agent processes this content.
  • Ingestion points: The SDK provides access to meeting roster data (ZoomToolSuiteAPI.Meeting.Participants), chat state, and live captions/transcription (references/package-and-api-map.md).
  • Capability inventory: The SDK includes broad controls for meeting management, including audio/video toggling, screen sharing, and participant roster controls (SKILL.md).
  • Boundary markers: The provided documentation and code snippets do not specify the use of delimiters or boundary markers when handling text-based data from meetings.
  • Sanitization: There are no instructions or examples regarding the sanitization or validation of meeting-derived strings (e.g., chat messages) before they are processed by an LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:55 PM
Security Audit — agent-trust-hub — zoom-plugin-sdk-macos