wind-mcp-skill

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests third-party media content via the financial_docs tools (see SKILL.md: "financial_docs.get_financial_news 获取来自第三方媒体的财经新闻报道") and the CLI’s vserver_financial_docs endpoint (scripts/cli.mjs), so the agent will read untrusted external news/articles that could contain instructions influencing subsequent actions.