sci-search
Warn
Audited by Snyk on May 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). This skill programmatically fetches and ingests content from open third‑party sources (arXiv via ArxivFetcher, PubMed via PubmedFetcher, and Web of Science via WoSFetcher in sci_search.py) and incorporates titles/abstracts/metadata into its workflow and output, exposing the agent to untrusted third‑party text that could influence subsequent decisions or responses.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata