load-balancing
Warn
Audited by Snyk on Jun 18, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). The required workflow ingests
input.csvprovided at runtime (step:python scripts/load_balancing_diagnostics.py input.csv --output diagnostics.json), and that CSV is free-form text authored by an outsider unless the operating user authored it; the script reads it viapd.read_csv(arguments.input_csv)and converts fields into LLM-readable JSON printed/returned.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata