๐ฅซ
Pantry
Docs
Hub
Backlog
Architecture
Use Cases
Mind Map
Sequences
Sprints
DS1
S1 Plan
Dev S1
Dev S2
S3 Plan
Dev S3
DS2
Dev S4
Landing V2
โ App
โ Spike
Pantry ยท Sequence Diagrams
Key system flows from sensor signal to user action
1
Auto Low-Stock Detection
Live โ Sprint 5
Sensor
Pi Hub
Edge Fn
Threshold Check
Supabase DB
React App
Lerato
MQTT: weight = 120g
POST /webhook {rice < threshold}
INSERT pending_item
ok: {item_id}
push notification
Rice is running low
Approve
UPDATE status: confirmed
โ Added to shopping list
UPDATE inventory_items.quantity
2
Receipt Scan โ Inventory Update
Sprint 3 โ Live
Lerato
React App
Edge Fn
Receipt Scanner
Cloud Vision
Google OCR
Supabase DB
upload receipt photo
POST /scan-receipt {image}
annotateImage()
{items, prices, store name}
parse + map items
items for review
confirm items
INSERT receipts + inventory_items
ok: {updated counts}
โ 8 items added to pantry
3
In-Store Shopping List
Phase 2 โ WhatsApp Bot
Thabo
WhatsApp Bot
Edge Fn
Shopping List
Supabase DB
"shopping list"
GET /shopping-list
SELECT active_items
{6 items}
formatted list
Rice, Milk, Eggs, Butter...
"tick rice"
PATCH /items/rice
UPDATE purchased = true
โ Rice marked as bought
NOT YET BUILT โ PHASE 2
โ solid: request/call
- - dashed: response/return
annotation: internal step