engineering-wechat-mini-program-developer

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions define a professional persona for WeChat development. There are no attempts to bypass safety filters, override agent instructions, or extract system prompts.
  • [DATA_EXFILTRATION]: No sensitive local files (such as SSH keys or environment variables) are accessed. Network operations are performed using standard WeChat APIs (wx.request) directed at illustrative placeholder domains (api.example.com).
  • [REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. It mentions established ecosystem tools like Taro and uni-app for cross-platform development, but does not perform unsafe package installations or dynamic code execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes handling external data via URL parameters (onLoad(options)), which is a standard functional requirement for Mini Program development. While this represents a potential attack surface, the context is strictly educational and architectural.
  • Ingestion points: Query parameters in the onLoad lifecycle method (e.g., pages/product/product.js).
  • Boundary markers: Not explicitly defined in the illustrative templates.
  • Capability inventory: Platform-standard APIs including wx.request, wx.setStorageSync, and wx.requestPayment.
  • Sanitization: The provided code snippets are templates for functionality and do not include complex validation logic, which is expected for this skill type.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 02:36 AM
Security Audit — agent-trust-hub — engineering-wechat-mini-program-developer