app.agragent.com
Satellite Maps
Integrate satellite remote sensing, climate analytics, crop data, computer vision, and AI-powered agronomic recommendations in a single platform — for any crop, anywhere.
A comprehensive suite of tools designed for researchers, agronomists, and farm managers to monitor and optimize crop production.
Sentinel-2 imagery via Google Earth Engine with 6 vegetation indices: NDVI, NDRE, MSAVI, TCARI, True Color, and False Color composites at 10m resolution.
Real-time climate data from Open-Meteo API: temperature, precipitation, humidity, solar radiation, evapotranspiration, GDD, chill hours, frost risk, and heat waves.
RNA-seq differential expression visualization with 3,603 DEGs across 8 phenological stages from 68 samples and multiple crop varieties.
Dataset integration with YOLO annotations for crop detection and counting. Over 300 annotated images for object detection in agricultural fields.
Machine learning models (Extra Trees Regressor) for yield estimation with confidence intervals and feature importance analysis.
Context-aware Claude-powered assistant providing agronomic recommendations based on real-time field, climate, and satellite data with 6 autonomous tools.
Upload your field boundaries, connect to satellite and climate data sources, and receive AI-powered recommendations.
Upload a KML/KMZ polygon or draw your field boundaries directly on the interactive map.
Automatically fetch Sentinel-2 imagery and historical climate data for your specific location and season.
Explore vegetation indices, climate KPIs, genomic data, and yield predictions through interactive charts and maps.
Consult the AI assistant for context-aware agronomic advice based on your field's real-time data.
Interactive geospatial mapping with polygon drawing, editing, and split-screen comparison.
Server-side satellite image processing with cloud masking and index computation.
10+ interactive charts for climate, genomic, and yield data visualization.
High-performance Python backend for AI assistant and data processing.
Agentic AI with tool-calling for context-aware agronomic recommendations.
Free, global historical weather data with no API key required.
Open-source PostgreSQL backend for data persistence, authentication, and real-time APIs.
State-of-the-art object detection model for crop identification and counting in field imagery.
| Source | Type | Access | Usage |
|---|---|---|---|
| Sentinel-2 SR Harmonized | Satellite imagery (10m) | Google Earth Engine | Vegetation indices, RGB composites |
| Open-Meteo Archive API | Historical weather | Free | Temperature, precipitation, humidity, radiation, ET₀ |
| WGISD (Embrapa) | Crop images (300+) | Public dataset | Object detection annotations |
| RNA-seq (Altimiras et al. 2024) | Genomic DEG data | Published | 3,603 DEGs across 8 phenological stages |
agragent is grounded in published scientific research in precision & digital agriculture from the Pontificia Universidad Católica de Valparaíso (PUCV), Chile.
Agronomy, Vol. 14(3), p. 613
DOI: 10.3390/agronomy14030613 →Progress in Artificial Intelligence (EPIA 2024), Springer LNCS Vol. 15400, Ch. 14
DOI: 10.1007/978-3-031-80084-9_14 →agragent is free, open source, and works with any crop and any location worldwide. No installation required.