Evaluation
AuraScore 79/100

Multi-Store Autonomous Replenishment Agent Evaluation Brief

Assess autonomous inventory replenishment agents on stockout prevention, overstock mitigation, and multi-echelon node balance.

Deploy this evaluation brief to benchmark autonomous replenishment and inventory rebalancing agents across distribution centers and retail stores. It pinpoints order volatility, bullwhip amplification, and supplier lead-time miscalculations.

Template

Role: Principal Supply Chain Automation Architect and Retail Operations Fellow specializing in multi-echelon autonomous inventory systems.

Context

  • Store network topology: {{fulfillment_network_topology}}
  • Target SKU demand dynamics: {{sku_demand_volatility}}
  • Stockout cost model: {{stockout_penalty_matrix}}
  • Vendor lead time reliability: {{vendor_lead_variance}}
  • Storage and handling caps: {{distribution_node_capacity}}
  • Agent reorder cycle: {{agent_inference_frequency}}

Task

Synthesize a rigorous technical evaluation brief assessing the automated inventory replenishment agent's purchase order generation, cross-echelon balancing decisions, and working capital risk across the store network.

Method

  1. Review purchase orders and node transfer recommendations generated by the agent across {{fulfillment_network_topology}}.
  2. Analyze agent handling of demand spikes in {{sku_demand_volatility}} to identify order over-amplification or bullwhip effects.
  3. Measure safety stock sizing and reorder triggers against actual historical vendor delivery swings defined in {{vendor_lead_variance}}.
  4. Audit recommended inbound transfer volumes against real physical warehouse bottlenecks in {{distribution_node_capacity}}.
  5. Calculate the trade-off balance between inventory carrying costs and service level penalties in {{stockout_penalty_matrix}}.
  6. Evaluate agent performance differences under varying decision intervals specified by {{agent_inference_frequency}}.
  7. Benchmark autonomous allocation against a standard (s, Q) continuous review replenishment baseline.
  8. Formulate operational boundary conditions to prevent phantom inventory over-ordering and trapped localized stock.

Constraints

  • The evaluation MUST quantify working capital inflation risk alongside stockout reduction metrics.
  • The brief MUST NOT recommend autonomous purchase order transmission directly to vendors without human-in-the-loop validation for volatile SKUs.
  • Recommendations MUST explicitly address receiving capacity limits at the store level during peak drop windows.
  • Avoid generic supply chain commentary; focus entirely on the agentic decision loop and constraint handling.

Output format

  • Replenishment Integrity Synthesis: Exactly 200 words evaluating autonomous order accuracy and network stability.
  • Echelon Performance Table: Matrix mapping regional DCs, hub stores, and spoke stores against fill rates and excess inventory build.
  • Failure Mode Diagnostics: 3 structured sections evaluating lead-time blindness, phantom inventory response, and capacity clipping.
  • System Governance Protocols: 4 mandatory technical controls required before expanding agent autonomous order thresholds.

Self-review

  • Confirm that the interaction between vendor lead time variance and warehouse capacity constraints is clearly examined.
  • Verify the presence of quantitative comparisons between carrying costs and stockout penalties.
  • Ensure all 6 context variables are actively integrated into the analytical steps.
AuraScore breakdown
79/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

ai-agents
agents-evaluation
retail-consumer-goods
retail
supply-chain
replenishment