zoom-mcp/team-chat
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: References official Zoom communication endpoints at mcp.zoom.us and documentation at developers.zoom.us.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external chat data, representing a potential surface for indirect prompt instructions.
- Ingestion points: Tools defined in references/tools.md for searching, filtering, and fetching chat messages and files.
- Boundary markers: Explicit safety rules in SKILL.md requiring user confirmation before the agent executes any write-capable chat tools.
- Capability inventory: Tools for sending messages, updating channels, and adding members.
- Sanitization: Mitigation is implemented via human-in-the-loop review guidelines provided in the skill instructions.
Audit Metadata