doomscroll
Warn
Audited by Snyk on Nov 19, 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 required workflow runs keyword searches across public X, TikTok, YouTube, Reddit, Hacker News, Product Hunt, Instagram and Threads, then reads the ranked top 60 rows and the top comments of the biggest posts, so any outsider's post or comment that matches a search is ingested without the agent selecting that specific item.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata