mx-financial-assistant
Fail
Audited by Snyk on Jun 15, 2026
Risk Level: HIGH
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.65). 脚本在运行时调用外部接口
https://ai-saas.eastmoney.com/.../assistant/ask,并将返回的data.displayData(模型回答的自由文本)直接作为answer输出进入后续LLM上下文/展示链路;该自由文本属于非操作用户/非本地可信来源的外部生成内容,存在间接提示注入风险。
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The script makes runtime POST requests to https://ai-saas.eastmoney.com/proxy/app-robo-advisor-api/assistant/ask and uses the returned JSON (data.displayData / references) to generate the agent's answers, so this external endpoint directly controls the agent's responses and is a required runtime dependency.
HIGH W008: Secret detected in skill content (API keys, tokens, passwords).
- Secret detected (high risk: 1.00). 在 scripts/generate_answer.py 中发现硬编码的 API key:EM_API_KEY 默认值为 "em_fjFqd4YB6Cqs52LF48XWbMDdLNq6MyNg"。该字符串为高熵、随机样式(前缀 em_ 加长随机字符),且被用于 HTTP 请求头 "em_api_key": EM_API_KEY 发送到 API_URL,符合“实际可用凭据”的定义,因此应视为真实凭据泄露。
其它可疑项(如 API_URL、TOOL_NAME、错误消息文本等)均为普通配置/文本,不构成秘密;也未发现私钥/PEM 区块或其他高熵凭据。没有明显的占位符或低强度示例密码需要忽略的情况。
(建议:移除该硬编码默认值,改为强制从环境变量注入或从安全秘钥管理服务读取,并视已暴露的 key 为泄露,进行轮换/撤销。)
Issues (3)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W008
HIGHSecret detected in skill content (API keys, tokens, passwords).
Audit Metadata