@magicmakerslab/geoverdant-mcp-server
v1.0.2
Published
GeoVerdant Model Context Protocol (MCP) server for farm soil metrics and satellite insights
Maintainers
Readme
@magicmakerslab/geoverdant-mcp-server
The official Model Context Protocol (MCP) server for GeoVerdant. This server allows AI agents (like Claude Desktop and Cursor) to interact with the GeoVerdant platform to manage farm projects, draw field boundaries (Areas of Interest), and request satellite analysis for soil metrics.
Quickstart (Claude Desktop)
Because this package is published to npm, you do not need to clone or install anything manually. You can configure Claude Desktop to run it directly via npx.
Open your Claude Desktop configuration file:
- Mac:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
- Mac:
Add the
geoverdant-mcpserver to your config:
{
"mcpServers": {
"geoverdant-mcp": {
"command": "npx",
"args": ["-y", "@magicmakerslab/geoverdant-mcp-server"]
}
}
}- Completely restart Claude Desktop.
- Open a new chat, and you should see the GeoVerdant tools available for the AI to use!
Available Tools
This MCP server exposes the following tools to the AI:
Organization & Project Management
list_organizations: Fetch organizations available to the user.create_project: Create a new farm project within an organization.list_projects: List all farm projects in an organization.get_project_details: Get detailed information about a specific project.
Field Boundaries (Areas of Interest)
create_aoi: Draw and define a new Area of Interest (AOI) boundary for a farm.list_aois: List all AOIs associated with a project.
Satellite & Soil Analysis
request_satellite_analysis: Trigger a new satellite analysis job for a specific AOI.get_analysis_status: Check the processing status of a requested satellite analysis.view_satellite_data: Retrieve the final soil metrics and satellite insights for a completed analysis.
Development & Deployment
If you are developing this server or want to host it in the cloud for web-based AI clients:
Local Development
Clone the repository and install dependencies:
npm install
npm run dev:cli # Run Stdio mode (for local Desktop testing)
npm run dev:cloud # Run Express/SSE mode (for Cloud testing)Publishing Updates
When publishing updates to npm, ensure you build the TypeScript source first:
npm run build
npm publish --access publicBuilt by the GeoVerdant Team @ MagicMakersLab
