cloud-functions

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security risks were identified. The skill functions as a documentation and utility guide for developers using Tencent CloudBase.
  • [PROMPT_INJECTION]: Deterministic detections regarding the concealment of actions were identified as false positives. The instructions (e.g., "Never echo that header or dump credential env vars to clients") are security best practices intended to guide the agent in writing secure code that protects environment secrets from exposure to external users, rather than an attempt to bypass agent safety guidelines.
  • [CREDENTIALS_UNSAFE]: The skill correctly implements credential safety protocols. It explicitly warns against hardcoding secrets and provides detailed instructions for using official platform tools (like manageAppAuth) to rotate and inject credentials via environment variables at runtime.
  • [EXTERNAL_DOWNLOADS]: All external references and domains identified (e.g., tcloudbasegateway.com, tencentyun.com, cloudbase.net) are official vendor resources belonging to Tencent CloudBase and are used for legitimate deployment and documentation purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 03:14 PM