RedBookSkills

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill shows and requires passing session/authentication tokens (e.g., xsec-token / XSEC_TOKEN) as CLI arguments and in commands (e.g., get-feed-detail, post-comment), which forces the agent to include secret token values verbatim in generated command output.

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

• Third-party content exposure detected (medium risk: 0.30). RedBookSkills 的 post-to-xhs 运行时会在“小红书内容检索与互动流程”中读取小红书页面/接口返回的免费文本（如搜索下拉推荐词、笔记详情与评论、通知 mentions），然后用于后续操作，因此外部用户可通过其在小红书上发布/评论内容来投喂该文本给代理链路。