# Geospatial Rule Grounding

> Skill: geospatial rule grounding

- Skill: `dingxingdi/geospatial-rule-grounding` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add dingxingdi/geospatial-rule-grounding`
- Raw SKILL.md: https://api.skillmd.com/api/skills/dingxingdi/geospatial-rule-grounding/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Data & Analytics
- Author: dingxingdi (https://skillmd.com/u/dingxingdi)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/dingxingdi/geospatial-rule-grounding

---


# Skill: geospatial rule grounding

## 1. Capability Definition & Real Case
* **Professional Definition**: The ability to translate operational and geographic natural language into precise PostGIS SQL predicates. This involves identifying spatial entities and their geometry types (Point, Line, Polygon), applying topological operators (ST_Intersects, ST_Within), managing Coordinate Reference Systems (ST_Transform), and differentiating between planar geometry and geodesic geography for accurate area and distance measurements.
* **Dimension Hierarchy**: Query Reasoning->Domain-Constrained Semantics->geospatial rule grounding

### Real Case
**[Case 1]**
* **Initial Environment**: A global historical climatology network (ghcn) table contains Point geometries (geom) representing weather stations, while an administrative 'states' table contains Polygon geometries (geom) for U.S. states.
* **Real Question**: Find all GHCN stations that intersect Pennsylvania.
* **Real Trajectory**: 1. Identify 'GHCN stations' as the source point table and 'Pennsylvania' as the target polygon within the states table. 2. Verify that both tables contain compatible geometry columns and identify ST_Intersects as the appropriate topological predicate. 3. Formulate an INNER JOIN between ghcn and states where states.name matches 'Pennsylvania'. 4. Project the station identifiers to produce the final list.
* **Real Answer**: SELECT ghcn.station_id, ghcn.name FROM ghcn JOIN states ON ST_Intersects(ghcn.geom, states.geom) WHERE states.name = 'Pennsylvania';
* **Why this demonstrates the capability**: This case demonstrates Point-in-Polygon spatial grounding. The agent must successfully navigate the join between two distinct spatial layers and apply a topological relationship (intersection) rather than a simple attribute match to find stations located within a specific geographic boundary.
---
**[Case 2]**
* **Initial Environment**: A database of protected areas (ne_protected_areas) contains a column 'unit_name' and a geometry column 'geom' in EPSG:4326. The user needs a precise metric in square kilometers.
* **Real Question**: What is the area of the protected area called 'Everglades' in square kilometers?
* **Real Trajectory**: 1. Search the protected areas table for the keyword 'Everglades' using case-insensitive matching. 2. Identify the geometry type as Polygon, making ST_Area a valid operation. 3. Recognize that the source CRS is degree-based (4326), necessitating a cast to the 'geography' type for a geodesic calculation in meters. 4. Apply a conversion factor (divide by 1,000,000) to transform the meter-based result into square kilometers.
* **Real Answer**: SELECT SUM(ST_Area(geom::geography)) / 1000000.0 AS area_sq_km FROM ne_protected_areas WHERE unit_name ILIKE '%Everglades%';
* **Why this demonstrates the capability**: This illustrates the critical distinction between planar and geodesic reasoning. A failure to cast to geography in a degrees-based CRS would yield a meaningless area value in square degrees, but the agent correctly uses geodesic arithmetic for metric accuracy.
---
**[Case 3]**
* **Initial Environment**: A census database contains a 'counties' table with per-county geometries and associated 'state' labels. The user requires a summary statistic based on the complexity of county boundaries.
* **Real Question**: Group counties by state and calculate the average perimeter in kilometers.
* **Real Trajectory**: 1. Map the 'perimeter' request to the ST_Perimeter function in PostGIS. 2. Use the 'geography' cast to obtain the perimeter in meters for the underlying EPSG:4326 geometries. 3. Group the dataset by the 'state' column to perform a categorical aggregation. 4. Calculate the average of the converted perimeters (meters / 1000) for each state grouping.
* **Real Answer**: SELECT state, AVG(ST_Perimeter(geom::geography) / 1000.0) AS avg_perimeter_km FROM counties GROUP BY state;
* **Why this demonstrates the capability**: This demonstrates advanced spatial aggregation. The agent must combine standard SQL grouping logic with spatial analysis functions, ensuring that the perimeter of individual polygons is correctly calculated and unit-converted before being passed to the AVG function.

## Pipeline Execution Instructions
To synthesize data for this capability, you must strictly follow a 3-phase pipeline. **Do not hallucinate steps.** Read the corresponding reference file for each phase sequentially:

1. **Phase 1: Environment Exploration**
   Read the exploration guidelines to discover raw knowledge seeds:
   `references/EXPLORATION.md`

2. **Phase 2: Trajectory Selection**
   Once Phase 1 is complete, read the selection criteria to evaluate the trajectory:
   `references/SELECTION.md`

3. **Phase 3: Data Synthesis**
   Once a trajectory passes Phase 2, read the synthesis instructions to generate the final data:
   `references/SYNTHESIS.md`