Return Processing Sla

Purpose

Calculates the time from return request creation to refund issuance for all completed returns in a period. Surfaces the average processing time, identifies orders that breached a configurable SLA threshold, and lists the longest-pending open returns. Read-only — no mutations.

Prerequisites

Parameters

Safety

> ℹ️ Read-only skill — no mutations are executed. Safe to run at any time.

Workflow Steps

Inputs: query: "created_at:>=''", first: 250, pagination cursor

Expected output: Returns with createdAt, status, refunds { createdAt }, order { name }

Inputs: Filter for orders with return_status:open to find pending returns exceeding SLA

Expected output: Open return orders with request dates

GraphQL Operations

# returns:query — validated against api_version 2025-01
query ReturnProcessingTimes($query: String!, $after: String) {
  returns(first: 250, after: $after, query: $query) {
    edges {
      node {
        id
        status
        createdAt
        order {
          id
          name
        }
        refunds(first: 3) {
          id
          createdAt
          totalRefundedSet {
            shopMoney {
              amount
              currencyCode
            }
          }
        }
        returnLineItems(first: 10) {
          edges {
            node {
              quantity
              returnReason
            }
          }
        }
      }
    }
    pageInfo {
      hasNextPage
      endCursor
    }
  }
}

# orders:query — validated against api_version 2025-01
query OrdersWithOpenReturns($query: String!, $after: String) {
  orders(first: 250, after: $after, query: $query) {
    edges {
      node {
        id
        name
        createdAt
        returnStatus
        returns(first: 5) {
          edges {
            node {
              id
              status
              createdAt
            }
          }
        }
      }
    }
    pageInfo {
      hasNextPage
      endCursor
    }
  }
}

Session Tracking

Claude MUST emit the following output at each stage. This is mandatory.

On start, emit:

╔══════════════════════════════════════════════╗
║  SKILL: Return Processing SLA                ║
║  Store: <store domain>                       ║
║  Started: <YYYY-MM-DD HH:MM UTC>             ║
╚══════════════════════════════════════════════╝

After each step, emit:

[N/TOTAL] <QUERY|MUTATION>  <OperationName>
          → Params: <brief summary of key inputs>
          → Result: <count or outcome>

On completion, emit:

For format: human (default):

══════════════════════════════════════════════
RETURN PROCESSING SLA  (<days_back> days, SLA: <sla_days> days)
  Returns analyzed:          <n>
  Avg processing time:       <d> days
  Within SLA (<sla_days>d):  <n>  (<pct>%)
  SLA breaches:              <n>  (<pct>%)
  Open returns pending:      <n>

  Longest open returns (no refund yet):
    Order <name>  — requested <n> days ago
  Output: return_sla_<date>.csv
══════════════════════════════════════════════

For format: json, emit:

{
  "skill": "return-processing-sla",
  "store": "<domain>",
  "period_days": 30,
  "sla_days": 5,
  "returns_analyzed": 0,
  "avg_processing_days": 0,
  "within_sla_count": 0,
  "sla_breach_count": 0,
  "open_pending_count": 0,
  "output_file": "return_sla_<date>.csv"
}

Output Format

CSV file return_sla_.csv with columns:

return_id, order_name, return_requested_at, refunded_at, processing_days, sla_breach, return_status

Error Handling

Best Practices