marketing-customer-research
Warn
Audited by Snyk on Aug 25, 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). Mode 2 “Go Find Research” directs the agent to gather and extract verbatim quotes from outsider-authored community/review/forum content (e.g., Reddit, G2, YouTube/TikTok comments), meaning runtime ingestion can include free text submitted by outsiders into those platforms.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata