Tewodros-Awraris/Farmer-Agent

AI agricultural advisory platform for Ethiopian farmers. Features multi-agent orchestration (Plan→Execute→Validate→Critic), 3 ML crop prediction models, 22+ MCP tools, multilingual support (Amharic/Oromo/English), semantic caching, and Human-in-the-Loop safety. Built with FastAPI & GPT-4o-mini. # FarmLine AI — Farmer Advisory & ML Prediction Agent FarmLine AI is an intelligent multi-agent agricultural advisory platform designed for Ethiopian farmers and agricultural extensions. It provides real-time personalized crop recommendations, optimal plantation timing predictions, harvest duration estimates, live weather forecasts, and interactive farm profile management. --- ## Key Features - **Multi-Agent Orchestrator Pipeline**: Deconstructs user agricultural queries through Planner, Router, Executor, Validator, and Critic agents. - **Machine Learning Prediction Suite**: - **Random Forest Crop Recommendation Model**: Recommends optimal crops based on 16 soil and climate feature parameters. - **Decision Tree Plantation Day Model**: Predicts exact recommended planting calendar days. - **Decision Tree Harvest Duration Model**: Estimates total days until harvest for target crops. - **Standalone Direct ML Predictor Page (`/predictor`)**: Simple, fast button-driven predictions with farm profile auto-fill. - **MCP Tool Server Integration**: Exposes prediction, weather, crop inventory, and location tools over standard Model Context Protocol. - **Interactive Parameter Form Gate (HITL)**: Promptly captures missing soil parameters (N, P, K, pH, rainfall, temperature) via interactive forms before running model inference. - **Farmer Dashboard (`/dashboard`)**: Direct visualization and editing of farm soil characteristics, climate metrics, location details, and active crop fields. --- ## Quick Start ### 1. Installation & Environment Setup ```bash # Clone the repository git clone git@github.com:Tewodros-Awraris/Farmer-Agent.git cd Farmer-Agent # Create virtual environment & install dependencies python -m venv venv source venv/bin/activate # On macOS/Linux # or: venv\Scripts\activate # On Windows pip install -r requirements.txt ``` ### 2. Run Application Server ```bash # Launch FastAPI development server python -m app.main ``` The server will start at `0.0.0.127. …