hap-api-website

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly directs extracting HAP Appkey/HAP-Sign and to place them into config.js and request headers (and to copy them from the dashboard/MCP URL), which requires the agent to handle and output secret values verbatim in code/requests.

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

• Third-party content exposure detected (medium risk: 0.30). 技能在运行时会通过 HAP 的 MCP 工具读取应用内工作表与字段结构（如 get_app_worksheets_list、get_worksheet_structure），这些属于租户/用户管理的数据源而非先行选择的“官方文档”内容；因此外部用户若能通过其上传/写入到 HAP 的文本字段影响这些结构/示例数据内容，LLM 会在结构获取阶段读取并可能受其提示注入影响。