agent-evaluation
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). SKILL.md describes an eval-building workflow that samples “real logged requests” and “real failures” as case inputs, which can include outsider-authored free text at runtime when those logs/failed requests are ingested to run and grade the agent.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata