light-review-rebuttal

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

• Third-party content exposure detected (high risk: 0.72). 该技能在“数据源”中要求运行 scripts/fetch_openreview.py 通过 OpenReview API 在运行时拉取他人（非操作用户）的公开审稿/评论文本（Official_Review/Official_Comment）并将其用于校准生成评审内容，因此存在“OUTSIDER-AUTHORED FREE TEXT”进入 LLM 上下文的间接提示注入风险。

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)