research-lit
Warn
Audited by Snyk on Nov 12, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). Outsider-authored arXiv, Semantic Scholar, Exa, OpenAlex, WebSearch, and Gemini results (titles, abstracts, excerpts) enter the context only after the agent runs a topic search. The attacker can publish poisoned papers or pages but cannot force them into a query's results, so this is low.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata