reference-verify
Warn
Audited by Snyk on Jun 22, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). The skill’s Step 3 “Verify Each Reference” uses WebSearch/WebFetch (e.g., fetching pages/DOIs from the public web) to verify titles/abstracts, so outsider-authored free text from runtime web pages is ingested into the LLM context for matching and claim support.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata