The Agent Skills Directory

[SAFE]: No security issues were detected. The skill is designed for local quantitative research and data summarization.
[INDIRECT_PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection via the ingestion of external data in scripts/dispersion_correlation_arbitrage_diagnostics.py.
Ingestion points: Data is read from a user-specified file path using pd.read_csv in scripts/dispersion_correlation_arbitrage_diagnostics.py.
Boundary markers: None are present to delimit the data from instructions.
Capability inventory: The script is limited to local file reading and writing of a JSON summary; it lacks network access or arbitrary command execution capabilities.
Sanitization: The script performs type-coercion using pd.to_numeric and pd.to_datetime with errors="coerce", which effectively strips non-numeric content from the diagnostic calculations. This results in a low-risk profile as the processed data is not re-interpreted as instructions.

dispersion-correlation-arbitrage