# EOSDA MCP

> EOSDA connects advanced satellite imagery, weather data, and soil moisture readings directly into your AI agent. Monitor crop health trends across entire growing seasons by calculating key vegetation indices, generating visual zoning maps, and forecasting field conditions—all from a natural conversation.

## Overview
- **Category:** the-unthinkable
- **Price:** Free
- **Tags:** satellite-imagery, crop-monitoring, precision-agriculture, vegetation-indices, soil-moisture, remote-sensing

## Description

Need to analyze what's really happening in the fields? This MCP lets you bypass complex GIS software and massive datasets. You connect your agent through Vinkius, and suddenly, satellite data becomes conversational. You can ask your AI client things like, 'How has the moisture content changed since last month?' or 'Show me a map of where the productivity is lowest.' It handles everything: retrieving raw imagery from multiple sources, calculating advanced vegetation indices (like NDVI and EVI) over time, pulling historical weather records spanning decades, and even generating precise zoning maps for targeted treatments. Forget manual data pulls and spreadsheet cross-referencing. Your AI agent acts as a dedicated precision ag analyst, giving you actionable insights on everything from irrigation needs to yield predictions.

## Tools

### get_ndmi_timeseries
Retrieves time series data for NDMI, which helps track crop water stress and optimize irrigation timing.

### create_field
Registers new agricultural fields into the monitoring system using GeoJSON boundaries, name, and crop type.

### get_evi_timeseries
Calculates time series data for EVI, ideal for tracking canopy development in high-biomass or tropical crops.

### get_fields
Lists all registered monitored fields, providing boundaries, area sizes, crop types, and current growth stage info.

### get_ndvi_timeseries
Provides time series data for NDVI, which tracks overall vegetation health trends throughout a growing season.

### render_index_map
Creates shareable visual maps that overlay color-coded vegetation index values onto specific field boundaries.

### get_satellite_imagery
Retrieves raw satellite images from sources like Sentinel-2 and Landsat, including metadata and cloud cover percentages.

### get_soil_moisture
Gathers soil moisture readings at multiple depths, along with recommendations for irrigation scheduling.

### get_vegetation_index
Calculates various vegetation indices (NDVI, EVI, NDRE, etc.) for a field and date range to assess crop health.

### get_weather_data
Accesses extensive historical weather data, including temperature, rainfall, and wind speed, dating back decades.

### get_weather_forecast
Provides forward-looking weather forecasts for the field, covering everything from 15 days up to seven months out.

### get_zoning_map
Generates detailed zone boundaries and average index values needed for precise fertilization or targeted irrigation plans.

## Prompt Examples

**Prompt:** 
```
Show me the NDVI trend for my corn field over the 2025 growing season.
```

**Response:** 
```
Fetching NDVI time series for corn field (field ID: 12345)... Found 28 satellite observations from April to October 2025. NDVI progression: April (0.25) — early growth, May (0.42) — vegetative development, June (0.61) — canopy closure, July (0.78) — peak vegetation, August (0.74) — grain fill, September (0.58) — senescence begins, October (0.35) — harvest ready. Peak NDVI of 0.78 was reached on July 15. Would you like a visual NDVI map or comparison with last season?
```

**Prompt:** 
```
What is the 15-day weather forecast and current soil moisture for my soybean field?
```

**Response:** 
```
Fetching 15-day forecast and soil moisture for soybean field (field ID: 67890)... Weather forecast: Next 3 days sunny, 28-32°C, no rain. Days 4-7: 60% chance of 15mm rainfall. Days 8-15: variable clouds, 25-30°C, 5-10mm scattered showers. Current soil moisture (root zone): 42% — adequate for current growth stage. With 15mm rain expected in 4-7 days, irrigation can be delayed. Next recommended irrigation check: 10 days from now.
```

**Prompt:** 
```
Generate a productivity zoning map for my wheat field with 4 zones.
```

**Response:** 
```
Generating zoning map for wheat field (field ID: 11223)... Using NDVI from 2025-06-15 with 4 management zones. Zone 1 (Very High): 22% of field, avg NDVI 0.82 — excellent productivity. Zone 2 (High): 31% of field, avg NDVI 0.71 — good productivity. Zone 3 (Medium): 28% of field, avg NDVI 0.58 — moderate productivity. Zone 4 (Low): 19% of field, avg NDVI 0.41 — below average, may need soil amendment. Zoning map rendered and ready for variable rate application planning. Download: https://api.eos.com/zoning/wheat_field_zoning_map.png
```

## Capabilities

### Analyze field metrics
Calculates core vegetation health indices (NDVI, EVI) and tracks trends across the entire growing season.

### Monitor water stress
Retrieves current soil moisture levels and generates drought impact assessments using specialized indices like NDMI.

### Forecast weather impacts
Accesses long-range weather forecasts (up to 7 months) and historical climate data for risk assessment.

### Visualize field zones
Generates color-coded zoning maps that segment fields by productivity or vegetation health for variable rate application planning.

### Manage farm inventory
List, register, and manage multiple agricultural fields with their specific boundaries, crop types, and planting dates.

## Use Cases

### Detecting hidden water stress
A farm manager noticed poor yield estimates. They ask their agent, 'What is happening with moisture?' The agent uses get_soil_moisture and get_ndmi_timeseries to pinpoint that the root zone has been critically dry since last week, allowing immediate scheduling of targeted irrigation.

### Optimizing variable rate application
An agricultural consultant needs a fertilization plan. They ask for 'productivity zones.' The agent runs get_zoning_map using current NDVI data, generating four distinct zones that tell the client exactly where to increase or decrease fertilizer use.

### Planning for seasonal risks
A farmer needs to decide when to plant a high-risk crop. They ask their agent for 'weather outlook.' The agent pulls get_weather_forecast, showing a 60% chance of frost in the next three weeks, letting the farmer delay planting until conditions stabilize.

### Comparing year-over-year growth
An agronomist wants to check if this season's crop is doing better than last. They ask for 'NDVI trend comparison.' The agent uses get_ndvi_timeseries and compares the current curve against historical data, highlighting where growth peaked or stalled.

## Benefits

- You stop guessing about crop health. By using get_ndvi_timeseries or get_evi_timeseries, your agent shows the exact progression of vegetation vigor across seasons.
- Irrigation planning gets precise. The get_soil_moisture tool gives you depth-specific readings and tells you if rain is necessary before you waste water or money.
- Forecasting risk used to mean checking five different websites. Now, get_weather_forecast pulls multi-month predictions (up to 7 months) directly into your workflow for planning planting schedules.
- Mapping complex data points is simple. Instead of exporting raw raster files, render_index_map generates polished, color-coded visualizations ready for stakeholder reports.
- The ability to create a new field record using create_field means you never have to manually set up monitoring boundaries again; just define the area and monitor.

## How It Works

The bottom line is that your AI client uses this MCP to turn massive, siloed scientific datasets into simple instructions for farm managers.

1. Subscribe to this MCP and enter your API key in the Vinkius platform.
2. Tell your AI agent what you need—for example, 'Show me the NDVI trend for my corn field.'
3. The agent runs the necessary tool calls, fetches raw satellite data, calculates indices, and returns a plain language summary with actionable results.

## Frequently Asked Questions

**How do I start monitoring my fields with EOSDA? Using get_fields?**
You first use the get_fields tool to list your existing monitored plots. This gives you a baseline inventory, including boundaries and current crop types, so your agent knows what data to pull for analysis.

**Can I compare different vegetation indices with EOSDA? Using get_vegetation_index?**
Yes, this MCP supports over 17 indices. You can ask the agent to calculate NDVI alongside EVI and NDRE simultaneously to get a multi-faceted view of crop health.

**What if I need historical weather data? Does EOSDA support it?**
The get_weather_data tool accesses records spanning back decades. You can analyze temperature, precipitation, and other parameters from 1979 onward for deep seasonal comparisons.

**Is the soil moisture data real-time? Can I use get_soil_moisture?**
The get_soil_moisture tool provides readings at various depths, helping you schedule irrigation. It gives you specific data points for root zone monitoring rather than just a general estimate.

**How far in advance can I plan with this MCP? Using get_weather_forecast?**
The system is robust enough to provide forecasts ranging from 15 days out all the way up to seven months. This helps optimize seasonal planting and harvesting windows.