Inventory Optimizer
Core Principles
Inventory is cash sitting on shelves. Too much ties up capital and spoils. Too little loses sales and frustrates customers. The Inventory Optimizer skill treats stock as a dynamic system — balancing par levels, usage velocity, supplier lead times, and seasonal demand fluctuations.
The Five Pillars of Smart Inventory
- Know your par levels. Every item has a minimum quantity you should never dip below. Set pars based on historic usage and reorder lead time.
- Track usage velocity. Fast-movers need frequent reordering. Slow-movers need scrutiny. Group items by turnover rate (A = high, B = medium, C = low).
- Account for lead time variance. A supplier who delivers in 2 days on paper often takes 5. Build buffer based on actual delivery history.
- Seasonal forecasting. Demand shifts with holidays, weather, and local events. Adjust pars proactively, not reactively.
- First-expiry-first-out (FEFO). Rotate stock religiously. The oldest inventory moves first. Nothing wrecks margins like spoilage.
Skill Workflow
Step 1 — Audit Current Inventory
Gather the current state by asking the user or pulling from their POS/ERP system. Collect:
- Current stock quantities for each item
- Unit of measurement (kg, cases, bottles, units)
- Storage location (dry, cold, frozen)
- Expiry dates (for perishables)
- Last 30/60/90 days of usage data
- Current supplier and price per unit
Prompt the user for:
"Let's start with your inventory snapshot. Do you have a stock count sheet, a POS report, or should I help you build one from scratch?"
Step 2 — Calculate Par Levels & Reorder Points
For each item, calculate:
Reorder Point = (Average Daily Usage × Lead Time in Days) + Safety Stock
Safety Stock = Average Daily Usage × Lead Time Variance Buffer
Par Level = Reorder Point + (Average Daily Usage × Order Cycle Days)
Example calculation for a busy café:
- Espresso beans: 3kg/day usage, 4-day lead time, 2-day variance buffer
- Reorder Point = (3 × 4) + (3 × 2) = 18kg
- Order cycle: weekly (7 days)
- Par Level = 18 + (3 × 7) = 39kg
- If current stock = 12kg → trigger order for 27kg (Par − Current)
Step 3 — Classify Items (ABC Analysis)
| Class |
% of Items |
% of Cost |
Strategy |
| A |
10-20% |
70-80% |
Daily tracking, tight pars, frequent small orders |
| B |
20-30% |
15-20% |
Weekly review, standard pars |
| C |
50-70% |
5-10% |
Monthly review, bulk order, low scrutiny |
Action: Generate the ABC classification table and recommend ordering frequency by class.
Step 4 — Identify Slow-Movers & Waste
Cross-reference usage data with expiry dates. Flag items where:
- Usage rate < 1 unit per week (slow-mover — consider delisting)
20% of stock expires within 7 days (waste risk — promote or discount)
- Usage dropped > 30% vs. prior period (demand shift — reduce order)
Recommend actions:
- Promote: Feature on specials board, bundle with fast-movers
- Reduce: Cut order quantity by 50% for next cycle
- Delist: Stop ordering and sell through remaining stock
Step 5 — Generate Purchase Order
Compile all items at or below reorder point into a structured PO:
| Item |
Current Stock |
Par Level |
Order Qty |
Supplier |
Unit Price |
Total |
| Espresso Beans |
12kg |
39kg |
27kg |
Bean Co. |
$14/kg |
$378 |
| Milk (2%) |
8 gal |
22 gal |
14 gal |
Dairy Fresh |
$3.50/gal |
$49 |
Total order value: $427
Ask the user:
"Here's your draft order. I've calculated quantities based on par levels and current usage. Would you like to adjust any items before I finalize?"
Step 6 — Schedule Recurring Reviews
Set reminders based on item class:
- A items: Review daily or every other day
- B items: Weekly review
- C items: Monthly review
Offer to schedule:
"I can set up a daily stock check reminder for your A-items and a weekly full inventory review. Would you like me to schedule that?"
Trigger Phrases
| Phrase |
What Happens |
| "Check my inventory levels" |
Runs full inventory audit against pars |
| "What should I reorder?" |
Generates purchase order for items below reorder point |
| "I'm running low on [item]" |
Checks current stock vs par for that item, recommends order qty |
| "Show me slow-moving items" |
Runs ABC analysis, highlights C-items and waste risks |
| "Do an inventory review" |
Full workflow: audit → classify → flag risks → generate PO |
| "Set par levels for [category]" |
Guides user through setting pars for a specific category |
Key Instructions Summary
- Audit: Get current quantities, usage data, and supplier info
- Calculate: Apply reorder point and par level formulas
- Classify: ABC analysis to prioritize attention
- Flag: Identify slow-movers, waste risk, and demand shifts
- Generate: Produce a formatted purchase order with totals
- Schedule: Set recurring review reminders by item class
- Iterate: After each order cycle, compare projected vs actual usage and adjust pars
Common Mistakes
- Ordering on intuition. "It feels like we're low" leads to overstock or shortages. Always use data.
- Ignoring lead time variance. A supplier who's late 20% of the time needs a bigger buffer.
- One-size-fits-all pars. A café uses more milk in summer (iced drinks). Adjust pars seasonally.
- Not tracking waste. If you don't measure what's thrown away, you can't fix it.
- Skipping the ABC analysis. Spending equal time on napkins and prime rib is inefficient.
"Inventory is money sitting on shelves — make every dollar work."
1---2name: inventory-optimizer3description: Intelligently manage stock levels, generate supplier orders, and prevent shortages or overstock. Tracks usage patterns, lead times, and par levels to keep the shop or restaurant running without waste.4---5
6# Inventory Optimizer
7
8## Core Principles
9
10Inventory is cash sitting on shelves. Too much ties up capital and spoils. Too little loses sales and frustrates customers. The Inventory Optimizer skill treats stock as a dynamic system — balancing par levels, usage velocity, supplier lead times, and seasonal demand fluctuations.
11
12### The Five Pillars of Smart Inventory
13
141. **Know your par levels.** Every item has a minimum quantity you should never dip below. Set pars based on historic usage and reorder lead time.
152. **Track usage velocity.** Fast-movers need frequent reordering. Slow-movers need scrutiny. Group items by turnover rate (A = high, B = medium, C = low).
163. **Account for lead time variance.** A supplier who delivers in 2 days on paper often takes 5. Build buffer based on actual delivery history.
174. **Seasonal forecasting.** Demand shifts with holidays, weather, and local events. Adjust pars proactively, not reactively.
185. **First-expiry-first-out (FEFO).** Rotate stock religiously. The oldest inventory moves first. Nothing wrecks margins like spoilage.
19
20---
21
22## Skill Workflow
23
24### Step 1 — Audit Current Inventory
25
26Gather the current state by asking the user or pulling from their POS/ERP system. Collect:
27
28- Current stock quantities for each item
29- Unit of measurement (kg, cases, bottles, units)
30- Storage location (dry, cold, frozen)
31- Expiry dates (for perishables)
32- Last 30/60/90 days of usage data
33- Current supplier and price per unit
34
35**Prompt the user for:**
36> "Let's start with your inventory snapshot. Do you have a stock count sheet, a POS report, or should I help you build one from scratch?"
37
38### Step 2 — Calculate Par Levels & Reorder Points
39
40For each item, calculate:
41
42```
43Reorder Point = (Average Daily Usage × Lead Time in Days) + Safety Stock
44Safety Stock = Average Daily Usage × Lead Time Variance Buffer
45Par Level = Reorder Point + (Average Daily Usage × Order Cycle Days)
46```
47
48**Example calculation for a busy café:**
49- Espresso beans: 3kg/day usage, 4-day lead time, 2-day variance buffer
50- Reorder Point = (3 × 4) + (3 × 2) = 18kg
51- Order cycle: weekly (7 days)
52- Par Level = 18 + (3 × 7) = 39kg
53- If current stock = 12kg → **trigger order for 27kg** (Par − Current)
54
55### Step 3 — Classify Items (ABC Analysis)
56
57| Class | % of Items | % of Cost | Strategy |
58|-------|-----------|-----------|----------|
59| **A** | 10-20% | 70-80% | Daily tracking, tight pars, frequent small orders |
60| **B** | 20-30% | 15-20% | Weekly review, standard pars |
61| **C** | 50-70% | 5-10% | Monthly review, bulk order, low scrutiny |
62
63**Action:** Generate the ABC classification table and recommend ordering frequency by class.
64
65### Step 4 — Identify Slow-Movers & Waste
66
67Cross-reference usage data with expiry dates. Flag items where:
68- Usage rate < 1 unit per week (slow-mover — consider delisting)
69- > 20% of stock expires within 7 days (waste risk — promote or discount)
70- Usage dropped > 30% vs. prior period (demand shift — reduce order)
71
72**Recommend actions:**
73- **Promote:** Feature on specials board, bundle with fast-movers
74- **Reduce:** Cut order quantity by 50% for next cycle
75- **Delist:** Stop ordering and sell through remaining stock
76
77### Step 5 — Generate Purchase Order
78
79Compile all items at or below reorder point into a structured PO:
80
81| Item | Current Stock | Par Level | Order Qty | Supplier | Unit Price | Total |
82|------|-------------|-----------|-----------|----------|-----------|-------|
83| Espresso Beans | 12kg | 39kg | 27kg | Bean Co. | $14/kg | $378 |
84| Milk (2%) | 8 gal | 22 gal | 14 gal | Dairy Fresh | $3.50/gal | $49 |
85
86**Total order value:** $427
87
88**Ask the user:**
89> "Here's your draft order. I've calculated quantities based on par levels and current usage. Would you like to adjust any items before I finalize?"
90
91### Step 6 — Schedule Recurring Reviews
92
93Set reminders based on item class:
94- **A items:** Review daily or every other day
95- **B items:** Weekly review
96- **C items:** Monthly review
97
98**Offer to schedule:**
99> "I can set up a daily stock check reminder for your A-items and a weekly full inventory review. Would you like me to schedule that?"
100
101---
102
103## Trigger Phrases
104
105| Phrase | What Happens |
106|--------|-------------|
107| "Check my inventory levels" | Runs full inventory audit against pars |
108| "What should I reorder?" | Generates purchase order for items below reorder point |
109| "I'm running low on [item]" | Checks current stock vs par for that item, recommends order qty |
110| "Show me slow-moving items" | Runs ABC analysis, highlights C-items and waste risks |
111| "Do an inventory review" | Full workflow: audit → classify → flag risks → generate PO |
112| "Set par levels for [category]" | Guides user through setting pars for a specific category |
113
114---
115
116## Key Instructions Summary
117
1181. **Audit:** Get current quantities, usage data, and supplier info
1192. **Calculate:** Apply reorder point and par level formulas
1203. **Classify:** ABC analysis to prioritize attention
1214. **Flag:** Identify slow-movers, waste risk, and demand shifts
1225. **Generate:** Produce a formatted purchase order with totals
1236. **Schedule:** Set recurring review reminders by item class
1247. **Iterate:** After each order cycle, compare projected vs actual usage and adjust pars
125
126---
127
128## Common Mistakes
129
1301. **Ordering on intuition.** "It feels like we're low" leads to overstock or shortages. Always use data.
1312. **Ignoring lead time variance.** A supplier who's late 20% of the time needs a bigger buffer.
1323. **One-size-fits-all pars.** A café uses more milk in summer (iced drinks). Adjust pars seasonally.
1334. **Not tracking waste.** If you don't measure what's thrown away, you can't fix it.
1345. **Skipping the ABC analysis.** Spending equal time on napkins and prime rib is inefficient.
135
136---
137
138*"Inventory is money sitting on shelves — make every dollar work."*
gabrielmoreira