Amanatal11/ethio-agri-privacy

A privacy-preserving multi-agent system using LangGraph and Differential Privacy to provide climate-resilient farming advice for Ethiopian smallholders. Supports Amharic and Afaan Oromoo. # Privacy-Preserving Multi-Agent Climate-Resilient Farming Advisor A research-grade multi-agent system designed for Ethiopian smallholder farmers, focusing on climate resilience and privacy-preserving federated learning simulation. ## Overview This project implements a multi-agent system designed to provide climate-resilient agricultural advice to Ethiopian smallholder farmers. It addresses the challenge of delivering personalized, data-driven insights while preserving the privacy of individual farm data. The system utilizes a federated learning simulation to aggregate regional insights without exposing raw user inputs. The architecture consists of five specialized agents orchestrated by **LangGraph**: 1. **Local Data Analyzer**: Securely processes user inputs locally. 2. **Federated Collaborator**: Aggregates insights using Differential Privacy. 3. **Crop/Weather Planner**: Generates climate-resilient recommendations. 4. **Privacy Auditor**: Monitors data flows for leaks. 5. **Synthesizer**: Compiles multilingual reports (English, Amharic, Afaan Oromoo). ## Target Audience - **Researchers**: Interested in Federated Learning and Multi-Agent Systems in Agriculture. - **Developers**: Building privacy-preserving AI applications. - **Agricultural Extension Workers**: Looking for tools to assist smallholder farmers. ## Prerequisites - **Python**: Version 3.10 or higher. - **API Keys**: OpenAI and Tavily API keys are required for agent functionality. - **OS**: Linux, macOS, or Windows (WSL recommended). ## Installation 1. **Clone the repository**: ```bash git clone github.com cd ethio-climate-agri-advisor ``` 2. **Create a virtual environment**: ```bash python3 -m venv venv source venv/bin/activate ``` 3. **Install dependencies**: ```bash pip install -e . # OR pip install -r requirements/base.txt ``` ## Environment Setup Create a `.env` file in the root directory with your API keys: ```env OPENAI_A …