technical-content-optimizer

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests and processes user-supplied blog drafts ("当用户提供博客草稿或技术内容时...")—untrusted, user-generated content that the agent must read and interpret as part of its workflow—so it could carry indirect prompt-injection payloads.