pain-point-marketing-loop

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs collecting and analyzing high-engagement user comments from public social platforms (e.g., 小红书, 知乎, 抖音, B站 in "Step 1: Pain Point Mining" and uses those comments as input to AI prompts), so it ingests untrusted, user-generated third‑party content for the agent to read and interpret.