How zuzith3art1st/updated-mvi-agri-agent works
Seed2Harvest is a Streamlit agricultural sales assistant for South African farmers. The user workflow is onboarding, then a bounded agricultural chat or product browsing, followed by basket preparation and a reservation quotation. The app also documents dialect normalization, inventory and transactional order persistence, delivery-location lookup, and payment/email aids. The excerpts show inventory retrieval and external-service calls, but do not show the reservation flow invoking the database transaction; that connection is therefore not asserted.
Main entry points: app.py Streamlit application · The_Database.py Inventory database
Components 15
Farmer experience
- Streamlit application
app.py - Client onboarding
app.py
Commerce and records
- Product catalogue
app.py - Inventory database
The_Database.py - Basket and totals
app.py - Order quotation
app.py - Order transactions
The_Database.py - Payment QR codes
app.py - Email handoff
app.py
Assistant and integrations
- Dialect mapping
app.py - Agricultural chat
app.py - Groq model
- Location lookup
app.py - Nominatim service
Other components
- Farmer
Connections 13
- Farmer → Streamlit application: uses
README.md - Streamlit application → Client onboarding: presents
app.py - Streamlit application → Agricultural chat: routes to
app.py - Agricultural chat → Dialect mapping: normalizes dialect
app.py - Agricultural chat → Groq model: requests completion
app.py - Streamlit application → Product catalogue: routes to
app.py - Streamlit application → Order quotation: routes to
app.py - Streamlit application → Inventory database: fetches inventory
app.py - Product catalogue → Basket and totals: updates quantities
app.py - Streamlit application → Location lookup: looks up address
app.py - Location lookup → Nominatim service: queries
app.py - Order quotation → Payment QR codes: generates codes
app.py - Order quotation → Email handoff: builds mail link
app.py