wechatpay-payment-integration

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). 该 Skill 的运行时会执行 scripts/wechatpay-docs-sync.py update 从远程 URL https://wx.gtimg.com/resource/wechatpay_api/wechatpay-docs.zip 下载并解压 Markdown 文档到本地 assets/微信支付官网文档/，随后在“文档检索与问答”中 Read 这些文档作为 LLM 上下文输入；该远程压缩包属于非操作用户/组织作者的外部来源。

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's mandatory preload runs scripts/wechatpay-docs-sync.py which at runtime downloads and extracts the remote documentation from https://wx.gtimg.com/resource/wechatpay_api/wechatpay-docs.zip, and those fetched docs are required and used as the knowledge base that directly control the agent's prompts/responses.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). 该 Skill 专门针对微信支付（支付网关）接入与运维，明确提供与微信支付 API（包括 APIv3）的示例代码、接口文档链接、接入质量评估及“APIv3 接口动态排障（查询支付单、退款单，协助排查查单失败）”等功能。它是专门为金融支付操作设计的技能，会定位和输出与支付/退款相关的接口与示例，属于直接针对支付网关的集成工具，因此具备对财务交易接口的直接执行/调用能力风险。