deep-research
Warn
Audited by Snyk on Aug 18, 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
bibliography_agent(Phase 2) ingests externally authored free text from search results / fetched records and possible user-provided corpus entries at runtime (e.g., abstracts/full-text used for screening and APA annotated bibliography), creating an indirect prompt-injection risk through attacker-controlled content.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata