Features • Quick Start • Installation • How It Works • Examples • Structure • Multi-Platform • Discovery

A professional Google Earth Engine skill pack for AI coding assistants (Codex, Claude Code, Cursor, Cline, Windsurf, Copilot, and more).
Write production-grade remote sensing code — zero experience required.

?? Author's GEE Blog: blog.csdn.net/qq_31988139 ? GEE tutorials & research papers

What is GEEPro?

GEEPro is a ready-to-use skill folder that teaches AI coding assistants how to write Google Earth Engine (GEE) code. Think of it as a plug-and-play brain for Codex or Claude — once installed, the AI can:

?? Every skill comes from real blog posts: Author's GEE Column (CSDN) ? Each code example is tested, published content, not generic tutorials.

Write Python scripts for Sentinel-2, Landsat, MODIS, Sentinel-1 SAR and more
Generate vegetation indices (NDVI, EVI, NBR, NDWI) automatically
Run machine learning classification (Random Forest, SVM, CART)
Detect floods, forest loss, urban growth, climate trends
Export results to Google Drive or local GeoTIFF files
Handle errors and optimize performance without you doing anything

No GEE experience? No problem. The AI handles everything. You just provide a Google Cloud Project ID and a study area.

Features

Category	What GEEPro Can Do
Satellite Data	Sentinel-2, Landsat 5/7/8/9, MODIS, Sentinel-1 SAR, VIIRS, ERA5, CHIRPS, Hansen, JRC Water
Vegetation	NDVI, EVI, NBR, NDWI, mNDWI — median composites, cloud-free mosaics
Land Cover	Random Forest classification, CART, SVM — with accuracy assessment
Water and Floods	SAR flood detection, JRC water occurrence, surface water change
Fire and Forest	dNBR burn severity, Hansen forest loss/gain, recovery ratios
Climate	ERA5-Land temperature trends, CHIRPS precipitation, MODIS LST
Agriculture	NDVI time series, harmonic phenology, crop type mapping
Urban	VIIRS nightlights, urban extent, change detection
Time Series	Linear trends, harmonic regression, LandTrendr, CCDC
Machine Learning	Random Forest, CART, SVM, KMeans clustering, accuracy metrics
Export	Google Drive, Earth Engine Asset, local GeoTIFF, tiled export for large areas
Error Handling	Automatic detection of 20+ common GEE errors with fix suggestions

Quick Start (for Beginners)

Step 1: Get a Google Cloud Project

Go to https://console.cloud.google.com/
Click Create Project (or select existing)
Enable Earth Engine API under APIs and Services
Go to https://signup.earthengine.google.com/ and verify your project

Step 2: Install GEEPro

# Windows (Codex)
git clone https://github.com/xingguangYan/GEE-pro.git "$env:USERPROFILE\\.codex\\skills\\GEEPro"

Windows (Claude Code)
git clone https://github.com/xingguangYan/GEE-pro.git "$env:USERPROFILE\.claude\skills\GEEPro"
macOS / Linux
git clone https://github.com/xingguangYan/GEE-pro.git ~/.codex/skills/GEEPro

Step 3: Authenticate Earth Engine (one-time)

pip install earthengine-api geemap geopandas
earthengine authenticate

Step 4: Use It!

Open Codex or Claude and say:

"Use GEEPro to calculate NDVI for my study area. My project ID is my-project-123."

The AI will:

Check your environment
Ask for your study area (or you can provide coordinates)
Write and explain the code
Help you export the results

Installation

Prerequisites

Requirement	How to Get It
Python 3.8+	Download from python.org
Earth Engine account	Register at earthengine.google.com
Google Cloud Project	Create at console.cloud.google.com
OpenAI Codex or Claude Code	Install the desktop app
Git	Download from git-scm.com

For OpenAI Codex (Windows PowerShell)

pip install earthengine-api geemap geopandas
earthengine authenticate
git clone https://github.com/xingguangYan/GEE-pro.git "$env:USERPROFILE\\.codex\\skills\\GEEPro"

For Claude Code (Windows PowerShell)

pip install earthengine-api geemap geopandas
earthengine authenticate
git clone https://github.com/xingguangYan/GEE-pro.git "$env:USERPROFILE\\.claude\\skills\\GEEPro"

macOS / Linux

pip install earthengine-api geemap geopandas
earthengine authenticate
git clone https://github.com/xingguangYan/GEE-pro.git ~/.codex/skills/GEEPro

How It Works

The GEEPro Workflow

When you ask the AI for something, it follows this process:

Clarify — Asks for your Project ID and study area (if not provided)
Check — Runs environment verification
Research — Consults 15 reference guides for best practices
Design — Determines scale, CRS, cloud masking strategy
Code — Writes complete, production-ready Python script
Confirm — Shows you the plan before running expensive computation
Export — Helps you save results to Google Drive or local disk

Conversation Examples

Beginner:
You: "Use GEEPro. My project is my-project. Show me NDVI for Beijing."
AI: "Checking environment... OK! Creating a Sentinel-2 NDVI composite for Beijing with 2024 imagery and cloud masking..."

Intermediate:
You: "Use GEEPro. Project: my-project. Run Random Forest classification on my shapefile."
AI: "Let me check your shapefile, then set up training samples with 70/30 split..."

Advanced:
You: "Use GEEPro. Project: my-project. Detect forest loss from 2020-2023 in Yunnan using Hansen."
AI: "Using Hansen global forest change dataset... Computing tree cover loss area..."

Examples

#	Example	Dataset	What You Learn
1	Sentinel-2 NDVI	COPERNICUS/S2_SR_HARMONIZED	Cloud masking, median composite, index calculation
2	Landsat Time Series	LANDSAT/LC08/C02/T1_L2	Annual composites, linear trend detection
3	Random Forest Classification	Sentinel-2 + training data	ML classifier, accuracy matrix, kappa
4	SAR Flood Mapping	COPERNICUS/S1_GRD	Speckle filtering, threshold detection
5	Urban Nightlights	NOAA/VIIRS/DNB	Urban extent detection, multi-year change
6	Climate Trends (ERA5)	ECMWF/ERA5_LAND/MONTHLY	Temperature trends, annual aggregation
7	Forest Change (Hansen)	UMD/hansen/global_forest_change	Loss/gain analysis, area statistics
8	Agriculture Phenology	MODIS/061/MCD43A4_NDVI	Harmonic regression, seasonal modeling
9	JRC Water Mapping	JRC/GSW1_4/GlobalSurfaceWater	Water occurrence, change types
10	Tiled Export	Sentinel-2	Large area partitioning, batch export

Each example includes: code.py (runnable), RUN.md (execution log), DATA_LAYER.md (dataset info), sources.md (references).

Project Structure

GEEPro/
  SKILL.md                     # Main AI instructions
  README.md                    # You are here
  requirements.txt             # Python packages to install
  LICENSE                      # MIT License

  references/                  # 15 expert reference guides
    01_local_environment.md    #   Python + Earth Engine setup
    02_network_proxy.md        #   Proxy for restricted networks
    03_research_design.md      #   Study area, scale, output planning
    04_data_selection.md       #   Choosing datasets + band semantics
    05_vector_roi.md           #   Local shapefiles, GeoJSON
    06_boundary_compute.md     #   ROI complexity + EECU management
    07_export_strategies.md    #   Drive, Asset, tiled, local export
    08_machine_learning.md     #   RF, SVM, CART, accuracy assessment
    09_time_series_analysis.md #   Linear trend, harmonic, LandTrendr
    10_advanced_reducers.md    #   reduceNeighborhood, zonal stats
    11_data_fusion.md          #   SAR+optical, thermal+optical
    12_common_errors.md        #   20+ GEE errors + diagnostic guides
    13_performance_tuning.md   #   EECU optimization strategies
    14_earth_engine_apps.md    #   Streamlit, ipywidgets, geemap Map
    15_js_api.md               #   JavaScript Code Editor patterns

  examples/                    # 10 ready-to-run workflows
  scripts/                     # Utility scripts
  gee_vector_db/               # Local knowledge database (30 MB)
  awesome-gee-community-datasets/  # Community dataset catalog
  templates/                   # Document templates
  platforms/                   # Multi-platform config files
    PLATFORMS.md               #   Overview of all platforms
    CLAUDE.md                  #   Claude Code instructions
    .cursorrules               #   Cursor rules
    .clinerules                #   Cline / Roo Code rules
    .windsurfrules             #   Windsurf rules
    .continuerules             #   Continue.dev rules
    AGENTS.md                  #   GitHub Copilot instructions
    mcp.json                   #   MCP client configuration

  agents/                      # AI assistant configuration
  assets/                      # Images and branding

Multi-Platform Support

GEEPro works across all major AI coding assistant platforms. See platforms/PLATFORMS.md for complete instructions.

Platform	Config File	How to Install
Codex (OpenAI)	`SKILL.md`	Copy to `~/.codex/skills/GEEPro/` or install via Plugins panel
Claude Code (Anthropic)	`CLAUDE.md`	Copy `platforms/CLAUDE.md` to project root, or install as skill
Cursor	`.cursorrules`	Copy `platforms/.cursorrules` to project root
Cline / Roo Code	`.clinerules`	Copy `platforms/.clinerules` to project root
Continue.dev	`.continuerules`	Copy `platforms/.continuerules` to project root
Windsurf	`.windsurfrules`	Copy `platforms/.windsurfrules` to project root
GitHub Copilot	`AGENTS.md`	Already in repository root
OpenAI GPTs	Custom GPT	Paste `SKILL.md` as GPT instructions + upload references
MCP Clients	`mcp.json`	Add `platforms/mcp.json` to MCP configuration

💡 One skill, all platforms. Install GEEPro once on your platform of choice and start writing production-grade Earth Engine code immediately.

🔍 How to Search & Discover GEEPro

GEEPro is designed to be easily discoverable across multiple channels. Here is how others can find it:

On GitHub

Search keywords: google earth engine, GEE skill, remote sensing AI, landsat skill, sentinel-2 skill, Earth Engine Codex, GEEPro
Topic tags: This repo is tagged with google-earth-engine, remote-sensing, codex-skill, claude-skill, geemap, earth-engine, landsat, sentinel, machine-learning, time-series-analysis
GitHub Explore: Search by topics on the GitHub Topics page

Via Package Managers

pip: pip install earthengine-api geemap geopandas (runtime dependencies)
Codex Plugin Marketplace: Search for "GEEPro" in the Codex Plugins panel (Personal marketplace)

Via Search Engines

Google: Search GEEPro earth engine skill codex or geepro remote sensing AI assistant
GitHub: Search GEE-pro earth engine skill or filter by topic:codex-skill

Auto-Discovery by AI Agents

When a user clones this repository or copies its config files, AI agents automatically discover GEEPro:

File	Discovered By	Auto-Trigger Keywords
`SKILL.md`	Codex (when in `skills/` dir)	`GEE`, `earth engine`, `geemap`, `landsat`, `sentinel-2`, `MODIS`, `NDVI`, `remote sensing`
`AGENTS.md`	GitHub Copilot, Codex, Cline	Any remote sensing or Earth Engine mention
`CLAUDE.md`	Claude Code (in project root)	Any GEE-related user request
`.cursorrules`	Cursor (in project root)	Any satellite/remote sensing topic
`.clinerules`	Cline / Roo Code (in project root)	Any Earth Engine workflow request
`.windsurfrules`	Windsurf (in project root)	Any geospatial question
`.continuerules`	Continue.dev (in project root)	Any remote sensing task

Recommended Search Queries

Share these queries to help others find GEEPro:

Search In	Query	Result
GitHub	`GEEPro earth engine codex skill`	github.com/xingguangYan/GEE-pro
GitHub Topics	`topic:codex-skill+topic:earth-engine`	Repos tagged with both topics
GitHub Topics	`topic:claude-skill+topic:remote-sensing`	Repos for Claude Code users
Google	`GEEPro remote sensing AI assistant codex claude`	README + community mentions
Codex (in-app)	Tell Codex: "Install GEEPro plugin"	Personal marketplace plugin

FAQ

Q: Do I need to know Python?
A: No. Just describe what you want in plain English, and the AI writes the code.

Q: Does this cost money to run?
A: Earth Engine is free for research and education. Exports to Google Drive are free.

Q: Can I use this with Claude Code?
A: Yes! Copy platforms/CLAUDE.md to your project root or clone to ~/.claude/skills/GEEPro/.

Q: Can I use this with Cursor, Cline, Windsurf, or Continue.dev?
A: Yes! Copy the corresponding config file from platforms/ to your project root. See the multi-platform guide.

Q: How do I search for this skill on GitHub?
A: Search "GEEPro earth engine codex skill" or filter by topics codex-skill and earth-engine.

Q: What if I am behind a firewall?
A: The skill includes proxy configuration guides (see references/02_network_proxy.md).

Q: How is this different from searching Google Earth Engine docs?
A: GEEPro packages expert knowledge, best practices, error handling, and complete workflows into one searchable skill.

License

GEEPro is released under the MIT License. See LICENSE for details.
Bundled third-party content retains its original license terms. See THIRD_PARTY_NOTICES.md.

Made with love for the Earth Engine community
GitHub • Discussions • Issues • Earth Engine • geemap

What is GEEPro?

Features

Quick Start (for Beginners)

Step 1: Get a Google Cloud Project

Step 2: Install GEEPro

Windows (Claude Code)

macOS / Linux

Step 3: Authenticate Earth Engine (one-time)

Step 4: Use It!

Installation

Prerequisites

For OpenAI Codex (Windows PowerShell)

For Claude Code (Windows PowerShell)

macOS / Linux

How It Works

The GEEPro Workflow

Conversation Examples

Examples

Project Structure

Multi-Platform Support

🔍 How to Search & Discover GEEPro

On GitHub

Via Package Managers

Via Search Engines

Auto-Discovery by AI Agents

Recommended Search Queries

FAQ

License

Reviews (0)