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
onLoadlifecycle 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, andwx.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