Skip to content
Back to search

earth-agent-chrome-ext

Earth Agent is a Cursor-like AI agent for Google Earth Engine. It can be used right in your browser as a Chrome extension or through MCP server support. It helps you do anything related to Google Earth Engine automatically through chatting (write code, run analysis, debug errors, explain maps, and manage your environment). Hatched from sundai.club.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/wybert--earth-agent-chrome-ext/badge.svg)](/listings/wybert--earth-agent-chrome-ext)

HTML

<a href="/listings/wybert--earth-agent-chrome-ext"><img src="/api/listings/wybert--earth-agent-chrome-ext/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Option 1: Install from Chrome Web Store (Easiest & Recommended)

**Best for**: Standard users who want to use OpenAI, Anthropic, or Google Gemini.

1. Visit the [Earth Agent page on the Chrome Web Store](https://chromewebstore.google.com/detail/earth-agent/hmpjiipbhhnppfdahieaafhdgdmhaple). 2. Click "Add to Chrome". 3. The extension will be added to your browser and will appear in your Chrome toolbar.

> [!NOTE] > The Web Store version does **not** support local connections (MCP Server or Ollama) due to store policies. For those features, use **Option 2**.

<details> <summary><b>Option 2: Download from GitHub Releases</b></summary>

1. Go to the [Releases page](https://github.com/wybert/earth-agent-chrome-ext/releases) 2. Download the latest `earth-agent-extension.zip` 3. Extract the zip file to a folder on your computer 4. Open Chrome and navigate to `chrome://extensions/` 5. Enable "Developer mode" (toggle in the top right) 6. Click "Load unpacked" and select the extracted folder 7. The extension will appear in your Chrome toolbar

</details>

<details> <summary><b>Option 3: Install from Source</b></summary>

1. Clone the repository 2. Install dependencies with `npm install` 3. Build the extension with `npm run build` 4. Load the unpacked extension from the `dist` directory in Chrome

</details>

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 318 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Earth Agent Chrome Extension
  2. <div align="center">
  3. <img src="src/assets/mydesign/Robot-earth-transparent-cut-edge.png" alt="Earth Agent Robot" width="200"/>
  4. </div>
  5. **Earth Agent** is a Cursor-like AI agent for Google Earth Engine. It can be used right in your browser as a Chrome extension or through MCP server support. It helps you do anything related to Google Earth Engine automatically through chatting (write code, run analysis, debug errors, explain maps, and manage your environment). Hatched from [sundai.club](https://www.sundai.club/projects/ad38a4e9-5cd5-4a90-b66c-c3f811cc5e8a).
  6. [![Watch the Demo](https://img.youtube.com/vi/RkSconBroyY/0.jpg)](https://youtu.be/RkSconBroyY?si=otXuZP6L8sp48K_1)
  7. > [!TIP]
  8. > **Quick Start**: Install from [Chrome Web Store](https://chromewebstore.google.com/detail/earth-agent/hmpjiipbhhnppfdahieaafhdgdmhaple), set up your API key (OpenAI/Anthropic/Google) or MCP server with Claude Code/Cursor/Zed/OpenCode, and say "Help me map NDVI for California".
  9. <details>
  10. <summary><b>Table of Contents</b></summary>
  11. - [Features](#features)
  12. - [Installation](#installation)
  13. - [MCP Server Support](#mcp-server-support-new)
  14. - [Configuration](#configuration)
  15. - [How to Use the Extension](#how-to-use-the-extension)
  16. - [Usage Examples & Prompts](#usage-examples--prompts)
  17. - [Available Tools & Functions](#available-tools--functions)
  18. - [Usage Tips](#usage-tips)
  19. - [Troubleshooting & Help](docs/TROUBLESHOOTING.md)
  20. - [Privacy and Security](#privacy-and-security)
  21. - [Roadmap](roadmap.md)
  22. - [Changelog](CHANGELOG.md)
  23. - [Contributing](#contributing)
  24. </details>
  25. ## Features
  26. - Chat interface for Earth Engine assistance
  27. - Knows Earth Engine Data Catalog as well as community dataset
  28. - Knows Earth Engine APIs
  29. - Help you write code, and run the code
  30. - Help debug the code
  31. - Help you explain the map
  32. - Planning and reasoning
  33. - Environment management tools (reset map/console, clear code)
  34. - Comprehensive agent testing framework with automated testing capabilities
  35. - **Custom Instructions**: Tailor the AI's behavior and responses to your specific needs. Work like AGENTS.MD.
  36. - **Agent Profiles**: Create and switch between different agent configurations for different tasks
  37. - **MCP Server Support**: Use Earth Agent tools directly from [Claude Code](https://claude.ai/code), [Cursor](https://cursor.sh), or [Zed](https://zed.dev) via the Model Context Protocol.
  38. ## Installation
  39. ### Option 1: Install from Chrome Web Store (Easiest & Recommended)
  40. **Best for**: Standard users who want to use OpenAI, Anthropic, or Google Gemini.
  41. 1. Visit the [Earth Agent page on the Chrome Web Store](https://chromewebstore.google.com/detail/earth-agent/hmpjiipbhhnppfdahieaafhdgdmhaple).
  42. 2. Click "Add to Chrome".
  43. 3. The extension will be added to your browser and will appear in your Chrome toolbar.
  44. > [!NOTE]
  45. > The 1.3.2 build includes opt-in local MCP and local-provider permissions. Public availability depends on publishing the matching connector and Chrome Web Store release; older store versions may not show this setup.
  46. <details>
  47. <summary><b>Option 2: Download from GitHub Releases</b></summary>
  48. 1. Go to the [Releases page](https://github.com/wybert/earth-agent-chrome-ext/releases)
  49. 2. Download the latest `earth-agent-extension.zip`
  50. 3. Extract the zip file to a folder on your computer
  51. 4. Open Chrome and navigate to `chrome://extensions/`
  52. 5. Enable "Developer mode" (toggle in the top right)
  53. 6. Click "Load unpacked" and select the extracted folder
  54. 7. The extension will appear in your Chrome toolbar
  55. </details>
  56. <details>
  57. <summary><b>Option 3: Install from Source</b></summary>
  58. 1. Clone the repository
  59. 2. Install dependencies with `npm install`
  60. 3. Build the extension with `npm run build`
  61. 4. Load the unpacked extension from the `dist` directory in Chrome
  62. </details>
  63. ## Configuration
  64. ### Experimental map coordinates and drawing
  65. The development branch includes geographic map context, coordinate-aware screenshots, Agent annotations, user-confirmed drawing, and Agent-owned native GEE geometries. Connect your map through **Settings → Map coordinates & drawing** first. See [setup, tool contracts, limitations and the real-GEE validation checklist](docs/map-spatial-interaction.md). This feature is not automatically enabled by installing the extension.
  66. > [!TIP]
  67. > You do not have to do this if you only want to use Earth Agent as MCP server. If you want to use it directly in your browser without external agents tools, you need to do this.
  68. ### API Keys
  69. 1. Click the Earth Agent extension icon in Chrome
  70. 2. Go to Settings
  71. 3. Configure your AI providers keys. we support OpenAI, Anthropic, Google, Z.AI, and Custom OpenAI-Compatible Providers. You need input at least one of them to use Earth Agent.
  72. 4. Go back to the chat page, select your preferred model
  73. 5. Start chatting with Earth Engine!
  74. <details>
  75. <summary><b>Custom Instructions (AGENTS.md)</b></summary>
  76. Just as software developers use files like `CLAUDE.md`, `GEMINI.md` or `AGENTS.md` to define project-specific rules for their AI coding assistants, Earth Agent allows you to set **Custom Instructions** to align the AI with your scientific or policy goals. Think of it as giving the agent a "Standard Operating Procedure" for your research. This is powerful for:
  77. - **Scientific Rigor**: e.g., "Always use the 'Cloud Score+' dataset for cloud masking in Sentinel-2 images."
  78. - **Policy & Communication**: e.g., "Draft all explanations as policy briefs suitable for decision-makers, avoiding jargon."
  79. - **Methodology Standardization**: e.g., "Always use median compositing instead of mean for annual composites."
  80. - **Coding Conventions**: e.g., "Use functional programming styles in GEE (e.g., `map()`) over client-side loops."
  81. **To configure:**
  82. 1. Open Extension Settings
  83. 2. Use **Custom Instructions** to set global preferences
  84. </details>
  85. <details>
  86. <summary><b>Agent Profiles (Custom Agents)</b></summary>
  87. This allows you to create your own agent with specific system prompts and tool access for specific workflows or use cases. Each agent profile can have its own:
  88. - **Custom System Prompt**: Define exactly how the agent should behave
  89. - **Tool Access**: Enable or disable specific tools for focused tasks
  90. **Example Profiles:**
  91. - **"Strict Coder"**: Focused purely on writing optimized code, with no chit-chat.
  92. - **"Tutor"**: Explains every step in detail for learning.
  93. - **"Analyst"**: Specialized in statistical analysis and visualization.
  94. **To configure:**
  95. 1. Open Extension Settings
  96. 2. Scroll to **Profiles** to create new agent personas
  97. </details>
  98. ## How to Use the Extension
  99. You can use this extension in two ways: through the MCP Server or inside the Earth Agent side panel. MCP servers expose the [same tools](##available-tools--functions) as the Earth Agent assistant. Check [MCP Server Support](#mcp-server-support-new) for details on using it as an MCP server. The following section will show you how to use it inside the Earth Agent side panel.
  100. ### Getting Started
  101. After configuration:
  102. 1. **Open Google Earth Engine Code Editor** in a Chrome tab: https://code.earthengine.google.com/
  103. 2. **Click the Earth Agent extension icon** in your Chrome toolbar
  104. 3. **Start chatting!** The agent can see your Earth Engine environment and help with any GEE-related tasks
  105. > [!NOTE]
  106. > The agent can only see the code in temp/new scripts in the Google Earth Engine Code Editor for now. You should save it once you think the script is ready.
  107. <details>
  108. <summary><b>Ask Mode vs Do Mode</b></summary>
  109. The agent operates in two modes that you can switch between:
  110. #### **Ask Mode (Read-Only)** 🔍
  111. Analysis and guidance mode - the agent can observe but not modify:
  112. - ✅ Read current code from the editor
  113. - ✅ Search documentation (geeDocs)
  114. - ✅ Take screenshots and inspect the page
  115. - ✅ Read console output and map state
  116. - ❌ Cannot insert or modify code
  117. - ❌ Cannot execute code
  118. - ❌ Cannot clear or reset the environment
  119. **Best for**: Learning, understanding code, exploring datasets, getting explanations
  120. #### **Do Mode (Full Access)** ⚡
  121. Full execution mode - the agent can take action:
  122. - ✅ All Ask Mode capabilities
  123. - ✅ Write and edit code in the editor
  124. - ✅ Execute code and monitor results
  125. - ✅ Clear map, console, and inspector
  126. - ✅ Complete end-to-end workflows
  127. **Best for**: Building scripts, debugging, automated analysis, multi-step tasks
  128. **Tip**: Start in Ask Mode when exploring new concepts, switch to Do Mode when ready to implement
  129. </details>
  130. ## Developer & Advanced Usage
  131. ### 🚀 Local MCP Server Support
  132. Connect a local MCP-capable Agent to your logged-in Earth Engine browser without an Earth Agent backend.
  133. 1. Install Node.js 22+ and extension version 1.3.2.
  134. 2. Open **Settings → Local MCP**, enable it and approve local access.
  135. 3. Use **Copy MCP setup** once. Merge the generated entry into your Agent's MCP settings and restart its connector. It contains only the extension ID and local port; no pairing token is required. The port setting is available under **Advanced connection settings** if needed.
  136. 4. Open Earth Engine, sign in, and ask your Agent to read the current editor code. You do not need an AI API key in the extension for MCP-only use.
  137. The Agent starts the connector automatically after configuration. MCP is opt-in; Settings shows connection status, GEE tab availability and retry controls.
  138. **Release note:** The generated configuration pins `earth-agent-mcp@1.3.2`. Publish that package before distributing this extension version. Until then, use the local-build configuration in the [MCP guide](./mcp-server/README.md). Web Store submission/review remains a separate release step.
  139. ## Usage Examples & Prompts
  140. 🌍 Getting Started
  141. ```
  142. Help me create a simple map showing NDVI for California using Landsat data
  143. ```
  144. 🛰️ Satellite Data Analysis
  145. ```
  146. Find and display the latest Sentinel-2 image over New York City with less than 10% cloud cover
  147. ```
  148. 📊 Data Processing (⚡ Requires Do Mode)
  149. ```
  150. Calculate zonal statistics for land use categories within protected areas and export the results to Drive
  151. ```
  152. 🚀 Complex Workflow (⚡ Requires Do Mode)
  153. ```
  154. I want to analyze deforestation in the Brazilian Amazon from 2000 to 2023.
  155. Please create a complete workflow that:
  156. 1. Loads appropriate satellite imagery
  157. 2. Calculates forest loss over time
  158. 3. Creates visualizations showing the changes
  159. 4. Generates statistics by state/region
  160. 5. Exports the results for further analysis
  161. ```
  162. 🔧 Debugging
  163. ```
  164. This code is running slowly, check my code for best practices and suggest improvements
  165. ```
  166. ## Available Tools & Functions
  167. Earth Agent provides a comprehensive set of tools available both in the Chrome Extension and via MCP.
  168. | Category | Agent Tool Name | MCP Tool Name | Description |
  169. | ---------------- | ----------------------------- | ------------------ | -------------------------------- |
  170. | **Code Editing** | `readCode` | `read_gee_code` | Read current code from editor |
  171. | | `editCode` | `edit_gee_code` | Edit code using search & replace |
  172. | | `writeCode` | `write_gee_code` | Overwrite entire editor content |
  173. | | `undoEdit` | `undo_gee_edit` | Revert last edit |
  174. | | `insertAtLine` | N/A | Insert text at specific line |
  175. | **Execution** | `runCurrentCode` | `run_gee_code` | Run the script in editor |
  176. | | `wait` | `wait` | Wait for execution/loading |
  177. | **Analysis** | `getConsoleOutput` | `gee_console` | Read console logs & errors |
  178. | | `getMapScreenPosition` | `gee_map_position` | Get map bounds & zoom |
  179. | | `getInspectorOutput` | `gee_inspector` | Read inspector panel data |
  180. | | `screenshot` | `gee_screenshot` | Take screenshot of interface |
  181. | | `snapshot` | `gee_snapshot` | Get accessibility DOM tree |
  182. | **Knowledge** | `geeDocs` | `gee_docs` | Search datasets & API docs |
  183. | | `weather` | `weather` | Get weather data |
  184. | | `dateTime` | `date_time` | Get current date/time |
  185. | **Control** | `clearMapInspectorAndConsole` | `clear_gee` | Reset workspace |
  186. | | `clickByRefId` | `click_element` | Click element by ID |
  187. | | `clickAtScreenPosition` | `click_position` | Click at coordinates |
  188. > [!NOTE]
  189. > Z.AI currently does not support multimodal inputs yet, so screenshot analysis features are unavailable when using Z.AI models. For the best Earth Engine integration experience with visual analysis capabilities, use OpenAI, Anthropic, or Google providers.
  190. ## Usage Tips
  191. ### **For Maximum Productivity**
  192. 1. **Use tool-compatible models** when you need automated code execution
  193. 2. **Be specific** in your requests for better results
  194. 3. **Ask for explanations** to learn Earth Engine concepts
  195. 4. **Use workspace management** tools to keep your environment clean
  196. 5. **Take screenshots** to document your work and results
  197. ### **Common Workflows**
  198. 1. **Exploratory Analysis**: Ask questions → Get code → Run and iterate
  199. 2. **Data Processing**: Define requirements → Generate pipeline → Execute and export
  200. 3. **Visualization**: Create map → Style layers → Add legends → Capture results
  201. 4. **Debugging**: Describe problem → Analyze code → Apply fixes → Test solution
  202. ## Troubleshooting & Help
  203. See **[Troubleshooting Guide](docs/TROUBLESHOOTING.md)** for solutions to common issues (API keys, connectivity, tools, screenshots) and help resources.
  204. ## Privacy and Security
  205. We take your privacy seriously. Your API keys are stored securely in your browser using Chrome Sync and never share to our serever (even if we have one). Chat history and code remain local, it will sent to the LLM provider you choose. For full details on data handling and security practices, please read our [Privacy Policy](PRIVACY_POLICY.md).
  206. ## Roadmap
  207. See [roadmap.md](roadmap.md) for our planned features and future development.
  208. ## Changelog
  209. See [CHANGELOG.md](CHANGELOG.md) for a detailed history of changes.
  210. ## Contributing
  211. We welcome contributions! Please see our [Contributing Guidelines](CONTRIBUTING.md) for details on how to set up your development environment, build the project, and submit pull requests.
  212. ## License
  213. [MIT](LICENSE)
  214. ## Thanks
  215. - [Sundai Club](https://www.sundai.club/)
  216. - React
  217. - Vercel AI SDK

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.