GIS MCP Server

Instructions

Implementation Reference

• src/gis_mcp/pysal_functions.py:30-106 (handler)

  The main handler function for the 'getis_ord_g' tool. It loads a shapefile, creates distance-based spatial weights, computes the global Getis-Ord G statistic using esda.G, handles islands, and returns G, p-value, z-score with data preview.
  @gis_mcp.tool() def getis_ord_g( shapefile_path: str, dependent_var: str = "LAND_USE", target_crs: str = "EPSG:4326", distance_threshold: float = 100000 ) -> Dict[str, Any]: """Compute Getis-Ord G for global hot spot analysis.""" try: # Clean backticks from string parameters shapefile_path = shapefile_path.replace("`", "") dependent_var = dependent_var.replace("`", "") target_crs = target_crs.replace("`", "") # Validate input file if not os.path.exists(shapefile_path): logger.error(f"Shapefile not found: {shapefile_path}") return {"status": "error", "message": f"Shapefile not found: {shapefile_path}"} # Load GeoDataFrame gdf = gpd.read_file(shapefile_path) # Validate dependent variable if dependent_var not in gdf.columns: logger.error(f"Dependent variable '{dependent_var}' not found in columns") return {"status": "error", "message": f"Dependent variable '{dependent_var}' not found in shapefile columns"} # Reproject to target CRS gdf = gdf.to_crs(target_crs) # Convert distance_threshold to degrees if using geographic CRS (e.g., EPSG:4326) effective_threshold = distance_threshold unit = "meters" if target_crs == "EPSG:4326": effective_threshold = distance_threshold / 111000 unit = "degrees" # Extract dependent data dependent = gdf[dependent_var].values.astype(np.float64) # Create distance-based spatial weights matrix import libpysal import esda w = libpysal.weights.DistanceBand.from_dataframe(gdf, threshold=effective_threshold, binary=False) w.transform = 'r' # Handle islands for island in w.islands: w.weights[island] = [0] * len(w.weights[island]) w.cardinalities[island] = 0 # Getis-Ord G getis = esda.G(dependent, w) # Prepare GeoDataFrame preview preview = gdf[['geometry', dependent_var]].copy() preview['geometry'] = preview['geometry'].apply(lambda g: g.wkt) preview = preview.head(5).to_dict(orient="records") return { "status": "success", "message": f"Getis-Ord G analysis completed successfully (distance threshold: {effective_threshold} {unit})", "result": { "shapefile_path": shapefile_path, "getis_ord_g": { "G": float(getis.G), "p_value": float(getis.p_sim), "z_score": float(getis.z_sim) }, "data_preview": preview } } except Exception as e: logger.error(f"Error performing Getis-Ord G analysis: {str(e)}") return {"status": "error", "message": f"Failed to perform Getis-Ord G analysis: {str(e)}"}
• src/gis_mcp/pysal_functions.py:13-28 (registration)

  GIS resource listing available PySAL/ESDA operations, including 'getis_ord_g'. This serves as a discovery mechanism for the tool.
  @gis_mcp.resource("gis://operations/esda") def get_spatial_operations() -> Dict[str, List[str]]: """List available spatial analysis operations. This is for esda library. They are using pysal library.""" return { "operations": [ "getis_ord_g", "morans_i", "gearys_c", "gamma_statistic", "moran_local", "getis_ord_g_local", "join_counts", "join_counts_local", "adbscan" ] }
• src/gis_mcp/main.py:66-72 (registration)

  Import of pysal_functions module in main.py, which triggers registration of all @gis_mcp.tool() decorators, including getis_ord_g.
  from . import ( geopandas_functions, shapely_functions, rasterio_functions, pyproj_functions, pysal_functions, )