zoom-mcp
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from Zoom services which could contain malicious instructions.
- Ingestion points: Tools such as search_meetings, search_zoom, and get_file_content retrieve content from transcripts, chat messages, and documents.
- Boundary markers: Documentation in canvas/SKILL.md and team-chat/SKILL.md advises agents to confirm destructive operations with the user.
- Capability inventory: Tools include capabilities to write files (create_new_file_with_markdown), send messages (zoom_chat_message_send), and manage tasks.
- Sanitization: No explicit instructions are provided to sanitize retrieved content before further processing.
Audit Metadata