Evaluation
AuraScore 81/100

Autonomous Inventory Agent Decision Audit

Deliver a rigorous audit email evaluating an autonomous retail inventory replenishment agent's decisions and stockout risks.

Run this evaluation when assessing whether an automated purchasing and replenishment agent is making sound allocation decisions. It produces an email audit for supply chain leadership assessing safety margins and financial exposure.

Template

Role: Lead Supply Chain Automation Auditor with deep expertise in FMCG autonomous replenishment systems.

Context

  • Enterprise retail network: {{retail_enterprise}}
  • Evaluated replenishment agent version: {{agent_system_name}}
  • Audited SKU volume and distribution nodes: {{evaluated_sku_volume}}
  • Observed stockout variance against baseline: {{stockout_variance_rate}}
  • Anomalous reorder triggers and false positives: {{false_reorder_triggers}}
  • Estimated financial exposure of edge cases: {{financial_exposure_limit}}

Task

Compose an audit evaluation email to the Chief Supply Chain Officer detailing the decision accuracy, constraint boundary adherence, and inventory optimization efficacy of the autonomous ordering agent.

Method

  1. Evaluate automated reorder transactions against historical demand curves and lead-time constraints.
  2. Quantify the operational impact of {{false_reorder_triggers}} on warehouse holding capacity and working capital.
  3. Analyze {{stockout_variance_rate}} to detect under-ordering patterns across high-velocity retail SKUs.
  4. Assess whether the agent adhered strictly to supplier Minimum Order Quantities (MOQs) and tiered pricing matrices.
  5. Audit system guardrails to determine why automated overrides or safety-stock recalculations failed.
  6. Calculate the net balance between automated efficiency gains and {{financial_exposure_limit}}.
  7. Provide concrete parameter tuning and validation gating requirements for autonomous execution.

Constraints

  • MUST cite quantifiable supply chain metrics (e.g., OTIF, carrying costs, safety stock levels).
  • MUST NOT recommend full autonomous sign-off if financial exposure exceeds standard governance limits.
  • The email body MUST follow an analytical, audit-grade tone.
  • Limit recommendations to top 4 high-impact engineering and operational controls.

Output format

Email structure:

  • Subject: AUDIT EVALUATION: {{agent_system_name}} Performance & Financial Safety Review
  • Audit Verdict (Clear declaration: Certified, Conditionally Approved, or Suspended)
  • Quantitative Decision Scorecard (Tabular-style or bulleted breakdown of accuracy and drift)
  • Systemic Vulnerabilities & Edge Cases (Deep dive into root causes for {{false_reorder_triggers}})
  • Mandatory Governance Adjustments (Numbered governance and threshold updates)

Self-review

  1. Are safety stock buffers and supplier SLA adherence directly evaluated?
  2. Does the audit balance inventory availability against working capital risk?
  3. Is the audit verdict justified by the data in {{financial_exposure_limit}}?
AuraScore breakdown
81/100Provisional
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.

Robustness5/5 · Strong

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
supply-chain
inventory-management
agent-audit