zoom-rest-api
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external Zoom API endpoints and webhooks, creating a surface for potential indirect prompt injection attacks where attacker-controlled data (e.g., meeting topics or chat messages) is processed by the agent.
- Ingestion points: External data enters the agent context through REST API responses and webhook payload objects described across multiple files including examples/webhook-server.md and references/meetings.md.
- Boundary markers: While data is typically handled in JSON format, the instructions do not specify delimiters or constraints to prevent the agent from acting on natural language instructions contained within the API data.
- Capability inventory: The skill provides an inventory of endpoints for significant platform operations including meeting lifecycle control, user management, and recording downloads.
- Sanitization: There is no evidence of explicit sanitization or filtering logic designed to neutralize instructional content within the ingested API data.
Audit Metadata