comment-mining
Warn
Audited by Snyk on Jun 23, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.95). The skill fetches and ingests public user-authored comment text from outsider platforms (e.g., TikTok/YouTube/Instagram/Facebook/Reddit/Rumble) via the listed runtime endpoints like
/v1/tiktok/video/comments, which becomes readable LLM context for classification and quoting.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata