context-engineering
Warn
Audited by Snyk on Jul 8, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.65). SKILL.md describes runtime “Dynamic Retrieval” from external vector stores (e.g., pgvector/Qdrant) and then injecting retrieved text into the “Compiler”/LLM payload, which can include outsider-authored free text if the underlying documents were not authored by the operating user.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata