miniprogram-development

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to recommend downloading the WeChat Developer Tools Nightly version from the official WeChat developers website (developers.weixin.qq.com). This is a standard procedure for accessing advanced developer features.
  • [COMMAND_EXECUTION]: The skill provides extensive documentation for using the wechatide command-line interface and miniprogram-ci. These tools are used for project initialization, compilation, simulator management, and deploying cloud functions. These are expected capabilities for a development-focused skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes instructions for implementing message push and customer service auto-reply features. This involves processing event data received from the WeChat platform (e.g., in references/message-push-customer-service.md).
  • Ingestion points: External message and event data from the WeChat platform are ingested via cloud function events.
  • Boundary markers: None identified in the provided templates.
  • Capability inventory: The skill has access to deployment tools (wechatide, miniprogram-ci) and cloud APIs (wx.cloud).
  • Sanitization: No specific sanitization or validation logic is provided for the incoming event data, relying on standard implementation practices.
  • [DYNAMIC_EXECUTION]: In references/cloudbase-integration.md, the skill suggests using npx @cloudbase/cloudbase-mcp@latest to run a Model Context Protocol server. This involves fetching and executing a Node.js package at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:59 PM
Security Audit — agent-trust-hub — miniprogram-development