get-review-theme

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts and parses webpages (SKILL.md: "URL：优先网页读取工具，失败则请用户提供正文" and README examples showing "从这个研究项目页面提取综述主题：https://example.com/project"), and the agent is expected to read and act on that arbitrary public web content to generate topics/keywords/core_questions that feed downstream actions, so untrusted third‑party content could influence behavior.