last30days-cn
Warn
Audited by Snyk on Oct 21, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The skill fetches external content from arbitrary GitHub issues/PRs and community platforms (Weibo, Xiaohongshu, Bilibili, Zhihu, Douyin, WeChat, Baidu, Toutiao, Hacker News, Reddit) where outsiders can post text that is subsequently ingested by the agent during research.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata