financial-computing-numpy
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external CSV data which represents a potential attack surface for indirect prompt injection.
- Ingestion points:
scripts/financial_computing_numpy_diagnostics.pyreads data from a user-provided CSV file usingpd.read_csv. - Boundary markers: No explicit boundary markers or instructions to ignore embedded content are used in the prompt logic relative to this data.
- Capability inventory: The script is limited to reading CSV files, performing numerical calculations (mean, quantiles, max drawdown), and writing a JSON summary to disk. It does not perform network operations or shell command execution.
- Sanitization: The script uses
pd.to_numericandpd.to_datetimewitherrors='coerce', which effectively filters out non-numeric or invalid date strings from the statistical calculations.
Audit Metadata