fund-risk-compare
Pass
Audited by Gen Agent Trust Hub on Nov 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external CSV files where column names and data are interpolated into reports without sanitization.
- Ingestion points:
scripts/etf_screener.py(viaread_csv_nav). - Boundary markers: No delimiters or ignore instructions are present in the output.
- Capability inventory: Local file read/write and mathematical operations. No network or subprocess capabilities detected.
- Sanitization: Input data is processed as raw strings or floats without validation for prompt injection patterns.
Audit Metadata