lark-workflow-social-tracker

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

• Third-party content exposure detected (high risk: 1.00). The workflow explicitly requires using "WebSearch" to fetch public social-media pages (e.g., "搜索公开可用的社交媒体数据（如 Social Blade）" and 模式二: 方式 A) and directs the AI to extract and act on that untrusted, user-generated web content for snapshots, analysis, and notifications, so third‑party page content can materially influence decisions and tool actions.