skills/zoom/skills/virtual-agent/web/Gen Agent Trust Hub

virtual-agent/web

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the official Zoom SDK hosted at https://us01ccistatic.zoom.us/us01cci/web-sdk/zcc-sdk.js. This is a standard resource for the Zoom Virtual Agent service.
  • [REMOTE_CODE_EXECUTION]: Examples in examples/campaign-and-entry-patterns.md demonstrate how to embed the vendor's JavaScript SDK. The execution is limited to the functionality provided by the official SDK on the user's web platform.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of user-specific data into the virtual agent context, which represents a standard attack surface for indirect injection.
  • Ingestion points: User context variables like firstName and email are defined in examples/campaign-and-entry-patterns.md and passed to the SDK.
  • Boundary markers: The provided code examples do not include explicit delimiters or guardrail instructions for user-supplied data.
  • Capability inventory: The SDK supports methods such as updateUserContext() and ChangeCampaign() as listed in concepts/lifecycle-and-events.md.
  • Sanitization: No sanitization logic is present in the documentation examples, as it is intended for demonstration of SDK usage.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:55 PM
Security Audit — agent-trust-hub — virtual-agent/web