deploy-prototype
Warn
Audited by Snyk on Aug 14, 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). In scripts/protolab.py, the
pairflow repeatedly reads free-text responses from the outsider-approved lab URL via/api/pair/{code}(and/api/pair,/api/health), and these responses are parsed and/or printed at runtime by the client.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata