MCP server that lets Claude run Google Earth Engine
An MCP server that lets Claude run Google Earth Engine.
Ask Claude to build a cloud-free satellite image of anywhere on Earth, compute vegetation indices, pull statistics, or classify land cover, and it actually does it on Google's servers. geeflow is the connector that makes that work: a Model Context Protocol (MCP) server exposing Earth Engine as a set of typed tools, plus a code tool so Claude can run arbitrary Earth Engine Python when a task needs something custom.
Underneath the MCP server is a full Python toolkit (usable on its own via CLI or import), so anything an AI session does, a human can reproduce.
Through Claude (or any MCP client), geeflow can:
- Find data: search the Earth Engine catalog (880+ datasets) and read dataset metadata - Build imagery: cloud-free, harmonized composites for any place and any year since 1972. It picks the right satellite for the year (Landsat MSS through Sentinel-2), cloud masks, renames bands to one common scheme, and falls back to backup sensors or wider date windows when scenes are scarce - Analyse: spectral indices (NDVI, SAVI, EVI, NDMI, NBR, NDWI, NDBI, red-edge), per-region statistics, scene availability - See it: PNG thumbnail links so you and Claude can look at the result - Training data: auto-generate labeled polygons from land-cover products like ESA WorldCover - Classify: a full land-cover classification workflow with honest, leakage-safe accuracy (optional, see below) - Export: send results to Google Drive or an Earth Engine asset, and track the jobs - Anything else: an executecode tool runs arbitrary Earth Engine Py
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/prahaladuk2208-arch--geeflow.html)
Девять российских маркетплейсов и китайский Taobao как MCP-серверы: Wildberries, Ozon, Яндекс Маркет, Детский мир, Авито, Мегамаркет, Lamoda, DNS, Ситилинк. Плюс сравнение цен по всем сразу. Только чт
korovin-aa97/talkthrough-mcpMCP server: turn narrated screen recordings into agent-ready data — local Whisper transcript, scene keyframes, OCR, wall-clock anchoring. Record your screen, talk — your AI agent files the bugs.
chris-asmussen/wardenOne MCP server that fronts many MCP servers + Skills behind a small tool set and routes to the best one, keeping an agent's context tiny. Migrate your Claude Code MCPs/skills behind it.
KernelLord/pickysteveSkill router and context picker for Claude Code, Cursor, and Aider. Auto-picks the right skill for every prompt, with prompt-injection filtering for MCP.
vinodborole/okf-kitTurn any website into a portable, agent-ready Open Knowledge Format (OKF) bundle — no LLM required. Crawl, sync, chat offline, and serve docs to coding agents over MCP.
AlexanderBoger/SaveContextGit LFS for your LLM context — 20x Context Savings — an MCP server that keeps big documents out of the model and returns byte-exact quotes for a fraction of the tokens.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.