unified-comment-analysis

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

• Third-party content exposure detected (high risk: 0.90). The skill's required 3-step workflow (described in SKILL.md and implemented in scripts/run.py) accepts user-provided public social-media URLs (抖音/小红书/B站/快手) and calls backend endpoints (parse-link -> create-task -> get-task) to fetch and analyze user-generated comments from those public platforms, which can materially influence returned suggestions and agent actions.