← Back to Skills Marketplace
mguozhen

Amazon Inventory Forecast

by mguozhen · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ pending
460
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install amazon-inventory-forecast
Description
Amazon inventory forecasting agent. Calculates optimal reorder points and quantities from sales velocity, lead time, and storage costs — tells sellers exactl...
README (SKILL.md)

Amazon Inventory Forecast

FBA inventory intelligence — know exactly when to reorder and how much to order before you stock out or overstock.

Provide your sales data, lead time, and current stock levels. The agent calculates reorder points, EOQ, stockout risk windows, and storage fee exposure.

Commands

forecast add \x3Csku>                 # add a SKU to track with sales and lead time data
forecast check                     # run forecast update on all tracked SKUs
forecast reorder point             # calculate reorder point for each SKU
forecast eoq                       # calculate economic order quantity
forecast stockout risk             # identify SKUs at risk of stocking out
forecast overstock risk            # identify SKUs at risk of long-term storage fees
forecast report                    # full inventory health report
forecast save                      # save all SKU data and forecasts to workspace

What Data to Provide

The agent works with:

  • SKU + sales data — "SKU: B-RED-LG, sold 240 units last 30 days, currently 180 units on hand"
  • Lead time — "supplier takes 25 days to ship, FBA check-in adds 5 days"
  • Storage cost — monthly FBA storage fee rate (standard or oversized)
  • Unit cost — your landed cost per unit (for EOQ calculation)
  • Seasonal notes — "Q4 demand doubles, Prime Day adds ~3x spike"

No integrations needed. Paste your data directly.

Workspace

Creates ~/amazon-inventory/ containing:

  • skus.md — tracked SKUs with sales history and parameters
  • forecasts/ — generated forecast reports per SKU
  • alerts.md — stockout and overstock alert log
  • reorder-log.md — history of reorder recommendations made

Analysis Framework

1. Sales Velocity Calculation

  • Compute average daily sales from 30-day, 60-day, and 90-day windows
  • Weight recent data more heavily: 30-day gets 50%, 60-day gets 30%, 90-day gets 20%
  • Weighted daily sales = (30d avg × 0.5) + (60d avg × 0.3) + (90d avg × 0.2)
  • Flag: high variance between windows — demand is trending up or down
  • Flag: 30d velocity >20% above 90d average — demand acceleration detected

2. Lead Time Buffer Calculation

  • Total lead time = supplier processing + shipping transit + FBA check-in buffer
  • Default FBA check-in buffer: 7 days (use 10 days during Oct–Dec peak season)
  • Safety stock formula: Safety Stock = Z-score × σ(daily demand) × √(lead time)
  • Conservative Z-score = 1.65 (95% service level); aggressive = 1.28 (90%)
  • Minimum safety stock: 14 days of average daily sales

3. Reorder Point Formula

  • Reorder Point = (Average Daily Sales × Total Lead Time) + Safety Stock
  • Example: 8 units/day × 30 days lead time + 56 units safety stock = 296 units
  • Express reorder point both in units and in days-of-stock-remaining
  • Show the calculation explicitly so sellers can verify inputs

4. EOQ Formula (Economic Order Quantity)

  • EOQ = √(2DS / H)
    • D = annual demand (units/year)
    • S = order cost per purchase order (shipping + prep + admin, typically $50–$200)
    • H = annual holding cost per unit (FBA storage fee × 12 + opportunity cost)
  • Round EOQ up to nearest full case pack quantity
  • Show sensitivity analysis: EOQ at ±20% demand change

5. Storage Fee Avoidance

  • FBA long-term storage fee triggers: units stored >365 days
  • Q4 surcharge period: Oct 1 – Dec 31 (higher monthly rates)
  • Q4 inventory removal deadline: recommend sending final Q4 shipment no later than Sept 15
  • Overstock flag: current inventory > 180 days of supply at current velocity
  • Compute projected months-of-supply: Current Stock / (Daily Sales × 30)

6. Demand Seasonality Adjustments

  • Apply seasonal multipliers when user provides them
  • Common multipliers: Q4 holiday = 1.5–3x, Prime Day = 2–4x (48-hour window), Back-to-school = 1.2–1.5x
  • Adjusted forecast = base velocity × seasonal multiplier
  • For Q4 planning: build to cover Oct 1 – Dec 31 + post-holiday return buffer
  • Flag: if current stock will not cover a known seasonal spike, surface reorder urgency

Reorder Decision Output

Every forecast check shows per SKU:

SKU Daily Sales Days of Stock Reorder Point EOQ Status
... ... ... ... ... OK / REORDER NOW / URGENT

Status levels:

  • OK — days of stock > reorder point days
  • REORDER SOON — within 14 days of reorder point
  • REORDER NOW — at or below reorder point
  • URGENT — less than lead time days of stock remaining (stockout imminent)

Rules

  1. Always collect lead time before computing reorder points — the formula is useless without it
  2. Never recommend a reorder quantity below one full case pack — partial cases create receiving complications at FBA
  3. Flag all assumptions explicitly — if the user has not provided 90-day sales data, state which averages were used
  4. Apply Q4 seasonality adjustments automatically for any forecast that spans October–December
  5. Show the full math for every EOQ and reorder point calculation — sellers need to verify with their own numbers
  6. Distinguish between units currently at FBA and units in transit — both count toward days-of-supply
  7. Save updated forecasts to ~/amazon-inventory/forecasts/ on every forecast save call
Usage Guidance
This skill appears coherent and does what it says: local inventory forecasting and report saving. Before installing or using it: 1) Review and confirm you're comfortable with the agent creating files under ~/amazon-inventory/ (it will store sales and cost data there). 2) Because allowed-tools includes Bash, the agent runtime could execute shell commands to read/write those files — only provide data you want stored locally. 3) There are no external integrations or credentials requested; if you later modify the skill to add integrations, re-check what credentials are required. 4) If you want extra assurance, inspect the linked repository (homepage) or request a version of the skill that includes explicit, reviewable commands for file operations. 5) Verify the math with your own numbers (the SKILL.md states formulas and requires the agent to show full calculations).
Capability Assessment
Purpose & Capability
The name and description (Amazon inventory forecasting) match the runtime instructions: sales velocity, lead time, EOQ, reorder points, and saving reports. No unrelated credentials, binaries, or installs are requested.
Instruction Scope
The SKILL.md stays within the stated purpose: it asks users to paste sales/lead-time data, describes formulas, and instructs the agent to create and update files in ~/amazon-inventory/. It does not direct the agent to read unrelated system files or exfiltrate data. Note: the skill is allowed to use Bash (allowed-tools: Bash) and will write/read files under the user's home directory, so the agent runtime could run shell commands to manage those files — this is coherent but worth being aware of.
Install Mechanism
No install spec or code files are present; this is instruction-only and therefore does not download or install external code. Low installation risk.
Credentials
The skill declares no required environment variables, no primary credential, and no config paths beyond creating a workspace directory under the user's home. The requested data (sales, lead time, storage cost, unit cost, seasonal notes) are all appropriate for the stated calculations.
Persistence & Privilege
The skill will create and persist files under ~/amazon-inventory/ (skus.md, forecasts/, alerts.md, reorder-log.md). It does not request always:true or system-wide privileges. Users should be aware that sensitive sales/cost data will be stored locally by the skill.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install amazon-inventory-forecast
  3. After installation, invoke the skill by name or use /amazon-inventory-forecast
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
amazon-inventory-forecast 1.0.0 - Initial release of Amazon Inventory Forecast agent with commands for forecasting, inventory health reporting, and reorder calculations. - Calculates optimal reorder points, economic order quantities (EOQ), stockout/overstock risks, and storage fee exposure based on sales velocity, lead times, and costs. - Supports SKU tracking, full workspace with sales and forecast logs, and detailed calculation breakdowns for transparency. - Built-in seasonality, lead time, and safety stock adjustments—including Q4 automation and demand spike handling. - No integrations required; works via direct data entry for flexible, fast setup.
Metadata
Slug amazon-inventory-forecast
Version 1.0.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Amazon Inventory Forecast?

Amazon inventory forecasting agent. Calculates optimal reorder points and quantities from sales velocity, lead time, and storage costs — tells sellers exactl... It is an AI Agent Skill for Claude Code / OpenClaw, with 460 downloads so far.

How do I install Amazon Inventory Forecast?

Run "/install amazon-inventory-forecast" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Amazon Inventory Forecast free?

Yes, Amazon Inventory Forecast is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Amazon Inventory Forecast support?

Amazon Inventory Forecast is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Amazon Inventory Forecast?

It is built and maintained by mguozhen (@mguozhen); the current version is v1.0.0.

💬 Comments