How does the gee agent skill improve my standard Google Earth Engine workflow?

This skill adds a "plan-first" layer to your workflow, creating editable Python scripts and validation reports so you can review and debug the geospatial logic before committing to a live Google Earth Engine execution.

Which AI agents or platforms is this skill compatible with?

The skill is designed for AI agents that support Python execution and file management; it includes SKILL.md instructions and Codex metadata to ensure seamless integration with your preferred agent environment.

Do I need to start a new project from scratch to use this toolkit?

No, the skill is designed to handle existing projects by diagnosing failures and continuing established workflows without needing to restart the entire analysis from scratch.

What specific files and assets are included in the download?

You receive the complete SKILL.md instruction set, Python CLI source code, editable workflow templates, a local knowledge base of GEE guidance, and a first-run setup guide.

Does the agent require manual confirmation before running live exports or scripts?

Yes, the toolkit includes preflight checks and an explicit confirmation step, allowing the agent to verify project selection and export destinations before any live tasks are initiated.

Do I need to be an expert in Python to use the gee agent skill?

A basic understanding of Python, raster resolution, and Google Earth Engine concepts (like Areas of Interest and export destinations) will help you get the most out of the automated recipes.

gee agent skill

Automate Google Earth Engine workflow planning, failure diagnosis, and evidence-backed geospatial execution.

Free

1 installSecurity scanned

Works with the AI tools you already use

+22 more

+18 agents

gee agent skill

Example session with this skill installed

Diagnose the export failure for project 'forest-cover-2023'. The last task ID was T56789 and it returned a 'Layer error: Image.reduceRegion: Too many pixels' message.

Read your context and instructions
Compiled the gee agent
Generated the document

I have analyzed the task failure for the forest-cover-2023 project. The diagnostic trace identifies a memory-related constraint during the reduction phase.

Diagnostic Report: Task T56789

Metric	Detail
Error Class	Resource Exhaustion (Too many pixels)
Fingerprint	`reduceRegion` at 30m resolution over AOI "Amazon_Basin_East"
Root Cause	The pixel count exceeds the default `maxPixels` threshold (1e9).
Status	Deterministic Failure
Recovery Strategy	Increment `maxPixels` or implement spatial sharding.

Proposed Fix: Updated Python Snippet


# Apply recovery: Increased maxPixels and tileScale for heavy reduction
stats = image.reduceRegion(
    reducer=ee.Reducer.mean(),
    geometry=roi,
    scale=30,
    maxPixels=1e13, # Increased from default
    tileScale=4     # Added to prevent memory overhead
)

Next steps

Validate the updated script using gee-skill plan --dry-run.
Update the gee-plan/v0.3 YAML with the new tileScale parameter.
Resubmit the task and monitor the export trace in the project ledger.

gee-agent-skill.pdf

PDF · document

Generated

Example file from a real run - the skill writes it into your workspace.

Connects securely to your tools. The creator never sees your data.

About this skill

Review your Earth Engine workflow before you run it!

GEE Workflow Toolkit helps remote-sensing researchers and geospatial developers turn supported Google Earth Engine requests into reviewable plans, editable Python scripts, and validation reports. Instead of treating generated code as a finished result, the toolkit keeps the workflow inspectable: clarify the task, retrieve relevant guidance, generate or adapt the script, validate it locally, and review the live-execution boundary.

What you can do?

Plan supported Earth Engine workflows using editable recipes.
Retrieve dataset, operator, and troubleshooting guidance from a bundled local knowledge base.
Generate Python scripts and check them before contacting Earth Engine.
Continue an established GEE project without unnecessarily restarting its workflow.
Diagnose observed failures while preserving completed work.
Use project selection, preflight checks, and explicit confirmation before live exports.

What you receive?

SKILL.md agent instructions and Codex metadata.
Python CLI source and editable workflow templates.
A recipe registry, local evidence cards, and a knowledge graph.
Public example inputs, evaluation resources, and technical documentation.
A first-run guide with installation instructions and starter requests.

Who it is for?

Researchers, GIS analysts, and geospatial developers who work with Earth Engine and want a reviewable, code-first workflow. Basic familiarity with Python, areas of interest, raster resolution, and export destinations is helpful.

How to install

Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.

~30 seconds

1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.