Smart Productivity Agent for Indian Agriculture
Built for AgentathonX 2026 - AI for Social Impact.
KrishiVanX is a Context-Aware Agricultural AI Agent designed to help farmers make data-driven decisions. Moving beyond standard AI chatbots, the KrishiVanX Agent autonomously retrieves real-time satellite weather data and local government policies to generate highly specific, actionable plans tailored to the farmer's exact environment.
- Real-Time Context Injection (Tool Use): The agent automatically fetches live weather metrics (temperature, humidity, rainfall, wind speed) via the Open-Meteo API based on the user's selected location before generating advice.
- Policy Awareness: Cross-references the user's location with a database of active state agricultural subsidies (e.g., UP Krishi Yantra Yojana, Maha-DBT) to recommend financial aid dynamically.
- Visual Diagnostics: Analyzes uploaded crop images to detect diseases and recommends organic/chemical remedies based on current weather conditions.
- Multilingual Voice Interface: Farmers can speak their queries in over 20 regional Indian languages, making the tool accessible to rural demographics.
- Cloud Memory: Automatically logs reports, images, and interaction history to an Azure Cosmos DB backend for persistent record-keeping.
- Frontend: Streamlit
- AI/LLM Engine: OpenAI (GPT-4o) via GitHub/Azure AI Models
- Database: Azure Cosmos DB
- External Tools/APIs: Open-Meteo API (Weather retrieval), SpeechRecognition (Audio input), gTTS (Text-to-Speech)
- Clone the repository:
git clone [https://github.com/ASbhay24/KrishivanX_2.0.git](https://github.com/ASbhay24/KrishivanX_2.0.git) cd KrishivanX_2.0