braintrust-elicit-eval-criteria
Warn
Audited by Snyk on Aug 20, 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’s runtime workflow describes eliciting criteria by running an elicitation session over “real traces” (including user turns and system outputs), which are outsider-authored free text embedded in monitored interaction data that the LLM judge reads to extract evaluation criteria.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata