ai-model-web

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill documents the use of the official @cloudbase/js-sdk and Tencent Cloud management APIs (via callCloudApi). All external resources and endpoints are legitimate vendor-owned infrastructure.
  • [SAFE]: The instructions include robust security practices, such as explicitly forbidding anonymous authentication for AI operations and directing users to manage credentials via environment variables and separate authentication-focused skills.
  • [SAFE]: A mandatory two-step preflight process is implemented to verify environment eligibility and model readiness before executing SDK code, preventing common runtime errors and unauthorized resource usage.
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection as it handles user-provided content in the messages array for text generation. However, this is an inherent and expected characteristic of AI model interfaces. The skill mitigates risk by enforcing authentication gates and providing clear implementation patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 06:19 AM