geospatial-analysis

The geospatial-analysis skill provides a robust framework for handling geographic data, specifically designed for agents working with Earth-science datasets, including earthquake records, plate boundaries, and complex spatial geometries. This skill ensures that agents avoid common pitfalls such as performing calculations on unprojected or incorrectly projected coordinate systems, which is critical for maintaining precision in scientific and engineering tasks. It is intended for developers, data scientists, and autonomous agents that need to perform high-fidelity spatial operations.

• Perform accurate spatial distance calculations by projecting data from Geographic Coordinate Systems (like EPSG:4326/WGS84) to metric-based Projected Coordinate Systems (like EPSG:4087).
• Execute efficient spatial filtering, such as determining if points fall within specific polygon boundaries using Shapely geometries and GeoPandas operations.
• Utilize advanced geometric operations including unary_union to simplify and combine complex feature collections like plate boundary segments.
• Handle diverse data inputs including GeoJSON files and raw coordinate arrays containing latitude and longitude, transforming them into structured GeoDataFrames.
• Optimize performance by filtering datasets before performing computationally expensive projection or spatial join operations.
• Access and manipulate geometry metadata to perform attribute-based filtering for specific regions, codes, or spatial clusters.
• Best practices for geometric integrity: agents should prioritize projecting once to a target metric CRS rather than repeated transformations within loops to maintain speed and data accuracy.
• The skill expects input data in standard geospatial formats and produces analytical outputs such as distance metrics in meters or kilometers, and filtered subsets of spatial data for further processing.
• Constraints: strictly requires valid coordinate metadata to define the starting CRS and assumes standard libraries like GeoPandas and Shapely are available in the runtime environment.