jamrynx/musika

AI market intelligence platform for Zimbabwe's informal agricultural economy. Features daily reference prices, buyer matching, a bilingual AI advisor, and pest warnings. It builds Zimbabwe's first structured sovereign dataset on the informal ag sector, directly serving the National AI Strategy's data agenda. # Musika — Agricultural Market Intelligence Platform (prototype) AI-powered price discovery, matching, and threat surveillance for Zimbabwe's informal agricultural economy. Built for the POTRAZ AI for Impact Challenge 2026. **All data in this prototype is labeled demonstration data**, calibrated to published market price ranges. Live collection via a price-agent network is a post-challenge funded milestone. ## Run the backend cd backend pip install -r requirements.txt python data/seed_demo_data.py musika.db # if musika.db not present uvicorn app.main:app --reload # docs at localhost ## Endpoints - GET /prices/current dashboard cards (15 headline products) - GET /prices/history explorer series + min-max band - GET /prices/spread one commodity across all markets - GET /prices/commodities|markets - GET/POST /listings, GET /listings/{id}/matches - POST /incidents, GET /incidents/map - GET /brief/today?lang=en|sn - POST /advisor/ask LLM-grounded (falls back to deterministic) ## Frontend (generate, don't hand-create) npx create-next-app@latest frontend --typescript --tailwind --app cd frontend && npm i recharts leaflet react-leaflet # set NEXT_PUBLIC_API_URL=localhost in .env.local