MCP 服务器

Planetary Computer MCP Server

A Python MCP server that provides unified access to satellite and geospatial data through natural language queries, with automatic place name geocoding and support for raster, vector, and Zarr formats.

README

Planetary Computer MCP Server

A Python implementation of the Planetary Computer MCP server, providing unified access to satellite and geospatial data through natural language queries.

Sample Outputs

<table> <tr> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/sentinel_2_l2a_alps.jpg" width="200"> Sentinel-2 Alps</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/sentinel_2_l2a_coastal-miami.jpg" width="200"> Sentinel-2 Miami</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/naip_small-seattle.jpg" width="200"> NAIP Seattle</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/naip_medium-la.jpg" width="200"> NAIP Los Angeles</td> </tr> <tr> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/hls2_l30_medium-la.jpg" width="200"> HLS L30 Los Angeles</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/modis_09A1_061_large-bay.jpg" width="200"> MODIS Bay Area</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/sentinel_1_rtc_coastal-miami.jpg" width="200"> Sentinel-1 SAR Miami</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/cop_dem_glo_30_coastal-miami.jpg" width="200"> Copernicus DEM Miami</td> </tr> <tr> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/esa_worldcover_alps.png" width="200"> ESA WorldCover Alps</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/io_lulc_annual_v02_rural-iowa.png" width="200"> IO LULC Iowa</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/ms-buildings.jpg" width="200"> MS Buildings Vector Data</td> <td align="center"><img src="https://raw.githubusercontent.com/isaaccorley/planetary-computer-mcp/main/assets/images/pet_preview.png" width="200"> TerraClimate PET Zarr Preview</td> </tr> </table>

<table> <tr> <td align="center"><img src="assets/images/gridmet-heatmap-animation.gif" width="300"> GridMET Climate Data Heatmap Animation</td> <td align="center"><img src="assets/images/terraclimate-heatmap-animation.gif" width="300"> TerraClimate Data Heatmap Animation</td> </tr> </table>

Features

Unified Interface: Single download_data tool that automatically detects datasets from natural language queries
Natural Language Geocoding: Automatically converts place names (e.g., "San Francisco", "the Alps", "Amazon rainforest") to geospatial bounding box coordinates using the Nominatim geocoding service—no need to manually specify coordinates
Multi-format Support: Raster (GeoTIFF), Vector (GeoParquet), and Zarr data
Automatic Visualization: Generate RGB/JPEG previews for LLM analysis
Fast Downloads: Uses odc-stac for efficient COG access

Installation

uv sync

Usage

As MCP Server

python -m planetary_computer_mcp.server

Direct API Usage

from planetary_computer_mcp.tools.download_data import download_data

# Download Sentinel-2 data for San Francisco
result = download_data(
    query="sentinel-2 imagery",
    aoi="San Francisco",
    time_range="2024-01-01/2024-01-31"
)

print(f"Raw data: {result['raw']}")
print(f"Visualization: {result['visualization']}")

Tools

download_data

Unified tool for raster, DEM, land cover, and climate data.

Parameters:

query: Natural language query (e.g., "sentinel-2", "elevation data")
aoi: Bounding box [W,S,E,N] or place name
time_range: ISO8601 datetime range
max_cloud_cover: Maximum cloud cover (optical data)

Returns:

Raw GeoTIFF/Zarr/Parquet file
RGB/JPEG visualization
Metadata

download_geometries

Tool for vector/building data.

Parameters:

collection: Collection ID (e.g., "ms-buildings")
aoi: Bounding box or place name
limit: Maximum features

Returns:

GeoParquet file
Map visualization
Feature count

Supported Datasets

See collections.md for the complete list of supported datasets.

Development

Setup

uv sync --dev

Testing

uv run pytest

Linting/Formatting

uv run pre-commit run --all-files

Architecture

src/
├── core/           # Core utilities
│   ├── stac_client.py    # STAC search wrapper
│   ├── geocoding.py      # Place name → bbox
│   ├── collections.py    # Dataset metadata
│   ├── raster_utils.py   # odc-stac helpers
│   ├── vector_utils.py   # DuckDB helpers
│   ├── visualization.py  # Matplotlib viz
│   └── zarr_utils.py     # Xarray Zarr helpers
├── tools/          # MCP tools
│   ├── download_data.py
│   └── download_geometries.py
└── server.py       # MCP server entry point