Evaluation
AuraScore 79/100

Dynamic Retail Pricing Agent Guardrail Evaluation Brief

Audit autonomous retail pricing agents against margin floors, brand equity rules, and competitor volatility shockwaves.

Apply this brief to evaluate dynamic repricing and markdown optimization agents before or during live market deployment. It identifies erratic repricing loops, margin erosion, and promotional rule conflicts across retail product catalogs.

Template

Role: Senior Director of Retail Yield Optimization and Algorithmic Trading Risk with extensive background in automated CPG pricing governance.

Context

  • Target merchandise category: {{retail_category_segment}}
  • Agent algorithm release: {{pricing_engine_build}}
  • Absolute margin floor: {{margin_floor_threshold}}
  • Market data refresh rate: {{competitor_feed_frequency}}
  • Catalog turnover status: {{inventory_aging_profile}}
  • Active promotional constraints: {{promotional_calendar_rules}}

Task

Produce an advanced algorithmic evaluation brief assessing the price elasticity decisions, margin safety guardrails, and market stability performance of the autonomous pricing agent across the target product catalog.

Method

  1. Analyze repricing telemetry from {{pricing_engine_build}} across {{retail_category_segment}} under volatile demand scenarios.
  2. Stress-test automated price changes against {{margin_floor_threshold}} to locate edge-case breaches caused by compounding discounts.
  3. Evaluate the agent's reaction curve to competitor price scraping anomalies at {{competitor_feed_frequency}} to detect race-to-the-bottom pricing loops.
  4. Measure inventory liquidation velocity in relation to {{inventory_aging_profile}} to confirm markdowns align with salvage targets.
  5. Check compatibility between autonomous base price fluctuations and scheduled promotional overrides in {{promotional_calendar_rules}}.
  6. Model gross margin dollar impact comparing the autonomous model against deterministic rule-based pricing baselines.
  7. Identify predatory scraping exposure and autonomous over-indexing on low-volume competitor stockouts.
  8. Establish definitive go/no-go criteria for deploying autonomous pricing to unconstrained live catalog tiers.

Constraints

  • The evaluation MUST verify that zero pricing decisions breached {{margin_floor_threshold}} under simulated synthetic market drops.
  • The brief MUST NOT endorse autonomous repricing for MAP (Minimum Advertised Price) protected SKUs without hardcoded floor stops.
  • All recommendations MUST preserve brand price integrity over short-term sales volume spikes.
  • Do not include general macroeconomic forecasts unrelated to catalog price execution.

Output format

  • Executive Scorecard: 1 summary table detailing 6 core risk dimensions (Margin, Compliance, Loop Risk, Velocity, Data Integrity, Stability).
  • Algorithmic Safety Analysis: 3 structured sub-sections examining floor breach risk, loop behavior, and promotional interference.
  • Margin and Revenue Impact Projection: 1 detailed comparison narrative with projected basis point deltas.
  • Deployment Directives: 5 bulleted mandate points specifying parameter limits for production operation.

Self-review

  • Ensure the relationship between promotional rules and dynamic base pricing is rigorously evaluated.
  • Confirm explicit verification of margin floor compliance across all edge scenarios.
  • Validate that no generic pricing advice is included and that all findings relate directly to agent behavior.
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
dynamic-pricing
pricing-agent