mcp · Global

geeflow

MCP server that lets Claude run Google Earth Engine

Claim this listing

Represent geeflow? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Data offered docs-verified: Earth Engine catalog search (880+ datasets) and dataset metadata · cloud-free harmonized composites (Landsat MSS through Sentinel-2) · spectral indices (NDVI, SAVI, EVI, NDMI, NBR, NDWI, NDBI, red-edge) · per-region statistics (mean, median, min-max, stdDev) · scene availability counts per sensor · auto-generated training polygons from land-cover products such as ESA WorldCover · arbitrary Earth Engine datasets via execute_code (climate, rainfall, temperature, terrain)
Coverage docs-verified: Any area on Earth · 1972 onwards (any year since 1972)
Formats llm-inferred: PNG · MCP stdio
Access docs-verified: Free and open source (MIT licence). Requires a Google Earth Engine account (free for research and non-commercial use) and a one-time earthengine authenticate step; credentials stay on the user's machine.

Use cases llm-inferred: Cloud-free dry-season NDVI/NDWI baseline for an ecology or BNG habitat assessment · Land-cover training polygons from ESA WorldCover for a site-scale classification · Area-wide vegetation and water-index statistics for a catchment or planning screening report
Audience llm-inferred: GIS analysts · ecologists · hydrologists · planners

Constraints llm-inferred: Requires Python 3.10+, a Google Earth Engine account and a Google Cloud project with the Earth Engine API enabled · Earth Engine access is free for research and non-commercial use; commercial consultancy use needs checking against Google's terms, which DOCS do not address · execute_code runs arbitrary Python on the local machine with the user's Earth Engine login; do not expose the server to untrusted clients · Initial earthengine authenticate step opens a browser · Roadmap features not yet available: Dynamic World and Hansen training labels, consensus logic, shapefile/KML/GPX/CSV import, local GeoTIFF backend, OpenStreetMap ingest, STAC backend, change-detection helpers · Land-cover classification workflow needs the optional lulc-engine package · No UK-specific datasets (e.g. river gauges, LiDAR DEM) are documented; these would need to be found in the Earth Engine catalog or added via execute_code · Spatial and temporal resolution and update frequency are not stated · Maintenance status and version history are not documented

Tools/install/auth auto-extracted from README; quality score heuristic (stars, recency, tool count, licence).

Metadata observed 15/10/2026 · research sweep: github-search