finflow

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to fetch and interpret real-time content from public third-party platforms (CLS, Jin10, Xueqiu, Gelonghui) via commands like "info flash"/"info news"/"topic show"/"quote f10"/"quote finance"/"market emotion", and Xueqiu in particular contains user-generated/untrusted content that the agent will read and use to drive decisions and outputs.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill instructs installing and running a third‑party CLI via "npm install -g @openduo/finflow", which fetches and executes remote code from the npm registry (e.g., registry.npmjs.org/@openduo/finflow) at runtime, so the fetched content can execute code locally.