A sophisticated demand forecasting and inventory optimization engine designed for multi-location retailers managing high-volume SKU assortments.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install inventory-demand-planning
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install inventory-demand-planning using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Inventory Demand Planning?
The Inventory Demand Planning skill codifies over 15 years of expert experience in retail supply chain management into an actionable framework for AI agents. It provides a robust set of logic for demand forecasting, safety stock calculation, and replenishment planning, specifically tailored for retailers operating between 40 to 200 stores. By implementing this within Openclaw Skills, businesses can bridge the gap between financial GMROI targets and warehouse execution.
This skill goes beyond simple arithmetic, offering deep technical implementations of Triple Exponential Smoothing (Holt-Winters), STL decomposition, and causal regression models. It is designed to handle the complexities of modern retail, including seasonal transitions, promotional lift estimation, and the management of vendor lead time variability, ensuring that inventory investment is always aligned with actual consumer demand.
Inventory Demand Planning Use Cases
- Automating weekly replenishment cycles for thousands of SKUs across regional distribution centers.
- Estimating promotional lift and post-promo demand dips to prevent excess stock after marketing campaigns.
- Performing automated ABC/XYZ classification to prioritize high-margin, predictable items for tighter inventory control.
- Managing seasonal buy commitments and markdown timing to maximize margin recovery.
- Identifying phantom inventory and supplier lead time shifts through automated tracking signals.
How Inventory Demand Planning Works
- The skill ingests POS data, ERP inventory positions, and vendor lead time history to establish a data baseline.
- It classifies every SKU using an ABC/XYZ matrix to determine the appropriate service level targets and review frequencies.
- A forecasting method is dynamically selected based on the demand pattern, ranging from moving averages for staples to Croston's method for intermittent demand.
- The engine calculates the optimal Inventory Position by accounting for on-hand, on-order, and committed quantities.
- Reorder logic is applied using EOQ or Min/Max levels, adjusted for vendor-specific constraints like Case Packs and Minimum Order Quantities.
- Automated escalation protocols are triggered if projected stockouts or forecast biases exceed defined thresholds.
Inventory Demand Planning Setup
To integrate this capability into your environment, follow these steps to initialize the skill within your Openclaw Skills workflow:
# Navigate to your local agent capability directory
cd capabilities/
# Clone the evos-capabilities repository
git clone https://github.com/evos-ai/evos-capabilities.git
# Register the inventory-demand-planning skill
openclaw register ./evos-capabilities/inventory-demand-planning
Inventory Demand Planning Data Schema & Taxonomy
The skill utilizes a comprehensive data schema to track inventory health and forecast accuracy:
| Attribute | Description | Metric Example |
|---|---|---|
| Forecast Accuracy | Measures the delta between plan and actuals | WMAPE, Bias, Tracking Signal |
| Inventory State | Real-time view of stock levels | On-Hand, On-Order, Backorders |
| Demand Patterns | Classification of sales behavior | Stable, Trending, Seasonal, Lumpy |
| Service Levels | Z-score based safety stock targets | 95% (Z=1.65), 99% (Z=2.33) |
| Promotion Metadata | Parameters for lift estimation | Promo Depth, Display Type, Cannibalization Factor |
Inventory Demand Planning Advanced Features
- Multi-model selection trees that automatically pivot between single, double, and triple exponential smoothing based on data volatility.
- Advanced safety stock formulas that incorporate both demand variability and vendor lead time coefficient of variation.
- Built-in markdown decision frameworks that trigger price actions based on sell-through velocity at season midpoints.
- Automated post-promo dip modeling to prevent over-ordering immediately following a high-volume event.
- Cross-category cannibalization logic to adjust forecasts when substitute items are promoted.
SKILL.md
Loading
Related Openclaw Skills
Customs & Trade Compliance
An AI-driven compliance specialist for automating HTS classification, duty optimization, and cross-border regulatory adherence.
Carrier Relationship Management
A codified expertise framework for managing carrier portfolios, negotiating freight rates, and automating logistics performance tracking.
Apollo Issue Review
A professional issue triage workflow that automates reproduction, evidence-checking, and drafting maintainer-grade GitHub responses for the Apollo ecosystem.
pi-speaker
A local audio playback workflow that enables AI agents to play speech and sound files directly through a Raspberry Pi or gateway host speaker.
Production Scheduling
Codified expertise for optimizing job sequencing, line balancing, and bottleneck resolution in manufacturing environments.
Returns & Reverse Logistics
A sophisticated AI agent skill for managing the end-to-end reverse logistics lifecycle, from RMA authorization to fraud detection and vendor recovery.
Featured*
