The Agent Skills Directory

[SAFE]: All components of the skill, including the main instructions and the supporting Python script, were analyzed and found to be free of security concerns. The script uses secure methods for file handling and scientific computation.- [INDIRECT_PROMPT_INJECTION]: The skill involves processing external hydrological data provided in CSV format.
Ingestion points: The scripts/groundwater_model.py script loads external CSV files for head, precipitation, evaporation, and pumping data using pandas.read_csv.
Boundary markers: Since the input is strictly numeric or date-based time series data for scientific modeling, no specific boundary markers for natural language instructions are used.
Capability inventory: Tool capabilities are constrained to file-based data ingestion, scientific model solving via the pastas library, and output generation (CSV, JSON, and PNG).
Sanitization: The skill relies on standard pandas and pastas validation logic for time series data. No additional sanitization or filtering is implemented or required for the primary scientific use case.

pastas