mcp-builder
Warn
Audited by Snyk on Aug 4, 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). The evaluation harness loads the evaluation XML (including the
<question>text authored by the user running the script), then sends it directly to the LLM inmessages=[{"role":"user","content": question}]at runtime, making outsider-authored free text part of the prompt.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata