An Agentic AI–powered web application that helps small and marginal farmers plan an entire agricultural season, manage multi-crop farms, optimize budgets, and receive weather risk alerts. The system dynamically updates farm plans based on changing conditions, acting as a virtual seasonal farm advisor.
Built as part of an Agentic AI Hackathon with a focus on ethical, explainable, and adaptable AI for real-world agriculture.
Small farmers often struggle to:
- Plan crops across an entire season
- Coordinate tasks like sowing, irrigation, and harvesting
- Anticipate weather risks (rain, heat, frost)
- Manage budgets and farm inputs efficiently
Lack of integrated planning tools leads to crop losses, wasted resources, and reduced income.
The Seasonal Farm Planner & Risk Alert Agent uses an agentic AI workflow to:
- Generate month-by-month crop plans using public crop calendars
- Suggest intercropping combinations
- Create detailed task schedules (sowing, irrigation, weeding, fertilization, harvest)
- Monitor weather forecasts (API or mock data)
- Trigger risk alerts and dynamically update plans when conditions change
The system shows what changed, why it changed, and how the plan adapts, making the AI transparent and explainable.
- Planning Agent – Builds a seasonal plan using crop calendars
- Task Generator – Creates date-wise farm activities
- Risk Monitor – Analyzes weather data for risks
- Plan Update Agent – Adjusts schedules based on detected risks
- Memory & Logs – Stores farm context and maintains plan update history
- 🌱 Multi-crop seasonal planning
- 📅 Month-by-month task calendar with dates
- 🚨 Weather risk alerts (heat, frost, heavy rain)
- 🔁 Automatic plan updates with change logs
- 💰 Budget estimation and input checklist
- 🌐 Mobile-friendly, responsive web interface
- 📄 Structured outputs (JSON/CSV ready)
- Location (district-level)
- Season (Kharif / Rabi)
- Crop selection (multi-crop)
- Field list (Field A, Field B, etc.)
- Budget constraints
- Weather forecast (API or mock data)
- Planner view (calendar and task tables)
- Weather risk alerts with explanations
- Updated plan history logs
- Budget and resource checklist
- Advisory-only system based on public agricultural guidelines
- No medical or chemical prescriptions beyond standard references
- Uses synthetic/demo data only
- No personal farmer data collected or shared
- Encourages consultation with local agricultural experts for high-risk scenarios
- FAO Crop Calendars
- ICAR Agricultural Guidelines
- IMD / OpenWeather (API or mock data)
- Local agricultural extension references
- Frontend: HTML, CSS, JavaScript (Bootstrap)
- Backend: Python (Flask)
- AI: Prompt-based Large Language Model integration
- Data: JSON files and mock APIs
- Python 3.9+
- Git
git clone https://github.com/your-username/seasonal-farm-planner.git
cd seasonal-farm-planner
pip install -r requirements.txt
python app.pyOpen your browser and go to:
http://127.0.0.1:5000
This project was developed for an Agentic AI Hackathon, emphasizing:
- Explainable AI decision-making
- Ethical and safe AI usage
- Practical agricultural impact
- Adaptability to real-world conditions
- Mobile application
- Offline / low-bandwidth mode
- SMS or WhatsApp alert integration
- IoT sensor integration (soil moisture, temperature)
- Expanded multilingual support (Tamil and other regional languages)
This application provides advisory guidance based on publicly available agricultural information. It is not a replacement for professional agricultural consultation.
Developed by [Your Name / Team Name] as part of a hackathon project.
⭐ If you find this project useful, please consider starring the repository!