Skip to content
Amazon Seller Skills v0.5.1

Monitor Amazon Refunds and Returns

Explain return volume, reasons, dispositions, concentration, listing-accuracy themes, and prior-period movement across FBA and merchant-fulfilled orders.

Analysis with pinned taxonomy
Example output visualization: Return causes
Return causes
skill output
Controllable
38%
Remorse
29%
Operations
24%
Other
9%
Primary input
FBA or MFN return records
Primary output
Return-rate and reason analysis
Powered by
Amazon Selling Partner MCP
Source contract
amazon-sp-refund-return-monitor
Is this the right Skill?

Use it for a clear operating job.

Best for
  • High-return ASIN diagnosis
  • Reason and disposition analysis
  • Prior-period return monitoring
Use another path when

Quoting a return rate when shipped-unit denominators are missing or incompatible.

Starter prompt

Start with the task, then add your scope.

Copy this into a connected client and replace the account, date, ASIN, or threshold details with yours.

Analyze our FBA and MFN returns for the last 90 days versus the prior period. Classify reasons, identify red-flag ASINs and listing-accuracy themes, and calculate return rate only where shipped units support it.
The Skill will request any required context or approval before guarded actions.
The outcome

A defined result, not a generic answer.

A reproducible returns report that separates controllable, remorse, operational, and other causes and identifies the ASINs that need action.

Inputs
  • FBA or MFN return records
  • Shipped units when available
  • Date range and comparison period
Outputs
  • Return-rate and reason analysis
  • Red-flag ASIN list
  • Markdown, workbook, or dashboard
Workflow

The playbook your agent follows.

  1. 01

    Acquire and inspect

    Load the return report from your agent-ready data lake through Agent Flow MCP, then inspect its actual fields and coverage.

  2. 02

    Normalize taxonomy

    Map reasons and dispositions into a pinned, auditable classification.

  3. 03

    Calculate carefully

    Compute concentration, deltas, and return rate only when a valid shipped-unit denominator exists.

  4. 04

    Render the deliverable

    Produce the requested report format without recomputing the analytical result.

Guardrails

What keeps the workflow dependable.

Return rate is never quoted without a valid denominator
Reason classifications remain reproducible
Listing-accuracy themes are evidence, not automatic listing edits
Connected systems

The context and tools it coordinates.

Amazon Agent Flow
Returns reports
Analysis runtime
amazon-sp-refund-return-monitor

Give your agent the complete playbook.

Connect Skills once, then start with the prompt above whenever this job comes up.