Use when choosing a coordinate reference system for world-scale or near-global geospatial work, including global maps, world choropleths, antimeridian issues, polar coverage, global grids, or deciding between EPSG:4326, EPSG:3857, Equal Earth, LAEA, Robinson, or other world projections.
Use when applying machine learning to geospatial data, including spatial feature engineering, morphometric indicators, urban analysis, spatial cross-validation, library selection, or integrating geopandas, momepy, pysal, osmnx, torchgeo, or rasterio into an ML pipeline.
Use when working with coordinate reference systems, EPSG codes, datum mismatches, reprojection choices, projected versus geographic CRS, unit confusion, or map alignment problems in geospatial data.
Use when inspecting a new geospatial dataset for the first time, including GeoJSON, Shapefile, GeoPackage, raster, CSV with coordinates, or mixed geo data. Helpful for schema review, CRS detection, geometry summary, dataset extent, and first-pass analysis planning.
Use when performing quality assurance or quality control on geo data before analysis, sharing, or publishing. Covers vector validity, attribute consistency, raster sanity checks, completeness, and basic geospatial data fitness review.
Use when working with PostGIS in PostgreSQL for geospatial SQL, including geometry or geography design, SRID handling, spatial indexing, ST_Intersects or ST_DWithin queries, raster or vector analysis patterns, and query tuning for PostGIS workloads.
Use when working with geospatial data in Snowflake, including GEOGRAPHY columns, WKT or GeoJSON loading, spatial joins, ST_INTERSECTS or ST_DISTANCE queries, geohash workflows, and performance-aware SQL design for geo analytics in Snowflake.
Use when ... Include the concrete geodata phrases a user would naturally say so the agent can discover the skill.