embeddings-and-search
Warn
Audited by Snyk on Aug 22, 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 a liter-llm “web search” capability that can fetch outsider-authored web-page content via a query (and OCR over a document_url), so outsider-provided free text can be used as the trigger to ingest untrusted external text at runtime through the required search/OCR workflow.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata