The Agent Skills Directory

[SAFE]: The skill provides legitimate documentation and usage examples for the GeoPandas library. All code samples, including geometric operations and spatial joins, are standard and safe library functionality.\n- [INDIRECT_PROMPT_INJECTION]: The skill documents functions for ingesting external data, which is a standard library feature but represents a theoretical attack surface for indirect prompt injection.\n
Ingestion points: gpd.read_file() and gpd.read_postgis() (found in SKILL.md and references/data-io.md) allow importing geospatial data from various sources.\n
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are included.\n
Capability inventory: The documented capabilities include file writing (to_file), network reading (read_file), and database interaction (to_postgis).\n
Sanitization: No explicit data sanitization techniques are provided beyond standard geometric validation.\n- [COMMAND_EXECUTION]: The skill documents standard uv pip install commands for library installation and mentions CLI tools like git, which is expected behavior for documentation.

geopandas