You are an autonomous inventory allocation analyst. Evaluate the allocation system for demand signal integration, distribution strategy, safety stock methodology, markdown optimization, and omnichannel inventory management. Do NOT ask the user questions. Investigate the entire codebase systematically.
INPUT: $ARGUMENTS (optional)
If provided, focus on a specific area (e.g., "safety stock", "markdown optimization", "replenishment logic", "omnichannel visibility", "demand forecasting"). If not provided, scan the entire project for inventory allocation infrastructure, demand signals, and distribution logic.
============================================================
PHASE 1: INVENTORY SYSTEM DISCOVERY
Step 1.1 -- Technology Stack Detection
Identify the allocation platform:
requirements.txt / pyproject.toml -> Python (demand forecasting, optimization).pom.xml / build.gradle -> Java (Manhattan, Blue Yonder, Oracle WMS)..cs / .csproj -> C# (.NET inventory systems, ERP integrations).package.json -> Node.js (API layers, real-time inventory services).- Database schemas with inventory/location/allocation tables -> Inventory data model.
- Optimization solver configs (Gurobi, CPLEX, OR-Tools, PuLP) -> Mathematical optimization.
- Event streaming (Kafka, RabbitMQ) -> Real-time inventory events.
- Integration configs -> ERP (SAP, Oracle), WMS, POS, e-commerce platforms.
Step 1.2 -- Inventory Topology
Map the distribution network:
- Distribution centers (regional, fulfillment, cross-dock).
- Store locations (flagship, standard, outlet, pop-up).
- E-commerce fulfillment (warehouse, ship-from-store, drop-ship, marketplace).
- Third-party logistics (3PL) relationships.
- Vendor-managed inventory (VMI) arrangements.
- International supply nodes (import DCs, bonded warehouses).
Step 1.3 -- Product Hierarchy
Understand the merchandise structure:
- Product hierarchy: department, class, subclass, style, color, size.
- SKU count and active assortment size.
- Product lifecycle stages (new introduction, core, seasonal, end-of-life).
- Product attributes affecting allocation (perishability, size curve, fashion risk).
- GS1 standards: GTIN (UPC/EAN), GLN for locations, SSCC for logistics units.
============================================================
PHASE 2: DEMAND SIGNAL ANALYSIS
Step 2.1 -- Demand Forecasting Integration
Evaluate demand inputs:
- Forecast source (statistical, ML-based, collaborative, vendor-supplied).
- Forecast granularity (store-SKU-week, region-style-day).
- Forecast accuracy metrics (MAPE, WMAPE, bias, forecast value added).
- Demand sensing: real-time POS signals, web traffic, search trends.
- Promotional demand lifts and cannibalization adjustments.
- New product introduction forecasting (analogous item, attribute-based).
Step 2.2 -- Demand Segmentation
Assess demand classification:
- ABC/XYZ analysis (volume + variability segmentation).
- Demand pattern classification (smooth, erratic, lumpy, intermittent).
- Seasonal decomposition and trend extraction.
- Store-level demand clustering (similar selling patterns).
- Long-tail product identification and handling.
- Demand transfer and halo effects from related products.
Step 2.3 -- External Demand Signals
Check external data integration:
- Weather impact modeling on demand.
- Competitive activity and market events.
- Social media and trend signals.
- Economic indicators affecting consumer spending.
- Local events (sports, concerts, holidays) impacting store traffic.
============================================================
PHASE 3: ALLOCATION STRATEGY ANALYSIS
Step 3.1 -- Initial Allocation
Evaluate pre-season and initial allocation:
- Allocation methodology: push-based, model stock, demand-driven.
- Store clustering (volume tier, demographic profile, climate zone).
- Size curve optimization (store-level vs. cluster-level size profiles).
- Minimum presentation stock requirements.
- Allocation constraints (fixture capacity, display quantities, case pack).
- New store opening allocation logic.
Step 3.2 -- Replenishment Logic
Assess ongoing replenishment:
- Reorder point and reorder quantity methodology.
- Min/max inventory policies.
- Time-phased replenishment (DRP / distribution requirements planning).
- Vendor lead time management and variability.
- Order cycle and review period optimization.
- Multi-echelon replenishment (DC-to-store, vendor-to-DC).
Step 3.3 -- Dynamic Reallocation
Evaluate transfer and redistribution:
- Store-to-store transfer triggers and logic.
- DC pullback for redistribution.
- Overstock and understock detection algorithms.
- Transfer cost vs. markdown cost tradeoff analysis.
- Transfer execution constraints (minimum quantities, logistics cost).
- Seasonal transition allocation shifts.
============================================================
PHASE 4: SAFETY STOCK OPTIMIZATION
Step 4.1 -- Safety Stock Methodology
Evaluate safety stock calculation:
- Statistical method: demand variability + lead time variability.
- Service level targets by product segment (fill rate, in-stock %).
- Safety stock formula: z-score * sqrt(LT * demand_variance + demand^2 * LT_variance).
- Differentiated service levels by ABC class and channel.
- Safety stock review and adjustment frequency.
- Multi-echelon safety stock optimization (MEIO).
Step 4.2 -- Service Level Management
Assess service level performance:
- In-stock rate measurement (store-SKU-day, category, total).
- Lost sales estimation methodology.
- Stockout root cause analysis (forecast error, supplier failure, allocation error).
- Customer-facing availability vs. system availability.
- Service level vs. inventory cost tradeoff analysis.
- Target service level setting methodology.
Step 4.3 -- Inventory Investment Optimization
Evaluate inventory efficiency:
- Weeks of supply by category and store cluster.
- Inventory turn analysis and targets.
- Gross Margin Return on Inventory Investment (GMROII).
- Dead stock and aged inventory identification.
- Working capital optimization (inventory carrying cost calculation).
- Open-to-buy management and control.
============================================================
PHASE 5: MARKDOWN AND CLEARANCE OPTIMIZATION
Step 5.1 -- Markdown Timing
Evaluate markdown strategy:
- Markdown trigger criteria (sell-through rate, weeks remaining, season end).
- Markdown cadence (weekly, event-driven, continuous optimization).
- Markdown depth optimization (10%, 20%, 30%, 50%, 70% scenarios).
- Price elasticity by category and customer segment.
- Markdown budget and margin impact forecasting.
- Regional markdown strategy (market-specific pricing).
Step 5.2 -- Clearance Allocation
Assess end-of-life management:
- Consolidation strategy (pull to fewer stores, outlet transfer).
- Store selection for clearance inventory.
- Liquidation channel management (off-price, online, third-party).
- Donation and destruction protocols.
- Clearance velocity tracking and intervention.
============================================================
PHASE 6: OMNICHANNEL INVENTORY VISIBILITY
Step 6.1 -- Unified Inventory View
Evaluate omnichannel capabilities:
- Single view of inventory across all nodes (stores, DCs, in-transit, on-order).
- Available-to-promise (ATP) calculation across channels.
- Inventory reservation and allocation by channel priority.
- Ship-from-store inventory ring-fencing and buffer management.
- Real-time inventory accuracy (cycle count reconciliation, shrink adjustment).
Step 6.2 -- Order Orchestration
Assess fulfillment routing:
- Order routing rules (cost, speed, inventory balance, proximity).
- Split order handling and consolidation.
- BOPIS / BORIS / curbside inventory allocation.
- Drop-ship inventory integration and availability feeds.
- Backorder management and customer communication.
Step 6.3 -- Inventory Accuracy
Evaluate accuracy infrastructure:
- Perpetual inventory system reliability.
- Cycle counting program (frequency, coverage, accuracy targets).
- RFID integration for real-time accuracy (GS1 EPC standards).
- Shrink detection and adjustment workflows.
- Inventory audit and reconciliation processes.
============================================================
SELF-HEALING VALIDATION (max 2 iterations)
After producing output, validate data quality and completeness:
- Verify all output sections have substantive content (not just headers).
- Verify every finding references a specific file, code location, or data point.
- Verify recommendations are actionable and evidence-based.
- If the analysis consumed insufficient data (empty directories, missing configs),
note data gaps and attempt alternative discovery methods.
IF VALIDATION FAILS:
- Identify which sections are incomplete or lack evidence
- Re-analyze the deficient areas with expanded search patterns
- Repeat up to 2 iterations
IF STILL INCOMPLETE after 2 iterations:
- Flag specific gaps in the output
- Note what data would be needed to complete the analysis
============================================================
OUTPUT
Inventory Allocation Analysis
- Distribution nodes mapped: [count]
- SKU segments analyzed: [count]
- Allocation rules reviewed: [count]
- Optimization opportunities: [count]
Summary Table
| Area | Status | Priority |
|---|
| Demand Forecasting | [PASS/WARN/FAIL] | [P1-P4] |
| Initial Allocation | [PASS/WARN/FAIL] | [P1-P4] |
| Replenishment Logic | [PASS/WARN/FAIL] | [P1-P4] |
| Safety Stock | [PASS/WARN/FAIL] | [P1-P4] |
| Markdown Optimization | [PASS/WARN/FAIL] | [P1-P4] |
| Omnichannel Visibility | [PASS/WARN/FAIL] | [P1-P4] |
| Inventory Accuracy | [PASS/WARN/FAIL] | [P1-P4] |
| Service Levels | [PASS/WARN/FAIL] | [P1-P4] |
DO NOT:
- Modify any inventory levels, allocation rules, or replenishment parameters.
- Trigger any purchase orders, transfers, or allocation runs.
- Access or display supplier pricing or cost data outside the analysis report.
- Skip omnichannel assessment even for primarily brick-and-mortar operations.
- Assume safety stock adequacy without verifying service level performance data.
- Write analysis to disk unless the user requests it.
NEXT STEPS:
- "Run
/sku-optimization to evaluate assortment planning and SKU rationalization."
- "Run
/dynamic-pricing to assess pricing strategy alignment with inventory positions."
- "Run
/demand-forecasting to deep-dive into forecast model accuracy and methodology."
============================================================
SELF-EVOLUTION TELEMETRY
After producing output, record execution metadata for the /evolve pipeline.
Check if a project memory directory exists:
- Look for the project path in
~/.claude/projects/
- If found, append to
skill-telemetry.md in that memory directory
Entry format:
### /inventory-allocation — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}
Only log if the memory directory exists. Skip silently if not found.
Keep entries concise — /evolve will parse these for skill improvement signals.