probabilistic-thinking
Warn
Audited by Snyk on Aug 22, 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 runtime workflow is a user-led probabilistic coaching “front door” (Elicit their real question → Base rate → Evidence → Update), so the agent ingests outsider-authored free text directly from the end-user without first selecting a specific trusted item.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata