Discount Hygiene Cleanup

Purpose

Audits the discount catalog for expired codes, codes with zero redemptions, and duplicate code prefixes. Discount sprawl accumulates over months of campaigns and makes the admin difficult to navigate. Optionally deletes flagged codes. Replaces manual discount cleanup and builds on the discount-ab-analysis skill with a write step.

Prerequisites

Parameters

Safety

> ⚠️ discountCodeDelete permanently removes discount codes. Deleted codes cannot be recovered. Customers who received a deleted code will find it invalid. Run with dry_run: true to review the flagged list before committing. Always check that expired codes are not referenced in active email campaigns before deleting.

Workflow Steps

Inputs: first: 250, select discount { ... on DiscountCodeBasic { codes, usageLimit, asyncUsageCount, endsAt, status } }, pagination cursor

Expected output: All discount codes with usage and expiry data; paginate until hasNextPage: false

Inputs: id:

Expected output: deletedCodeDiscountId, userErrors

GraphQL Operations

# discountNodes:query — validated against api_version 2025-01
query DiscountAudit($after: String) {
  discountNodes(first: 250, after: $after) {
    edges {
      node {
        id
        discount {
          ... on DiscountCodeBasic {
            title
            status
            createdAt
            endsAt
            asyncUsageCount
            usageLimit
            codes(first: 5) {
              edges {
                node {
                  id
                  code
                }
              }
            }
          }
          ... on DiscountCodeBxgy {
            title
            status
            createdAt
            endsAt
            asyncUsageCount
            usageLimit
          }
          ... on DiscountCodeFreeShipping {
            title
            status
            createdAt
            endsAt
            asyncUsageCount
            usageLimit
          }
        }
      }
    }
    pageInfo {
      hasNextPage
      endCursor
    }
  }
}

# discountCodeDelete:mutation — validated against api_version 2025-01
mutation DiscountCodeDelete($id: ID!) {
  discountCodeDelete(id: $id) {
    deletedCodeDiscountId
    userErrors {
      field
      message
    }
  }
}

Session Tracking

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

On start, emit:

╔══════════════════════════════════════════════╗
║  SKILL: Discount Hygiene Cleanup             ║
║  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>

If dry_run: true, prefix every mutation step with [DRY RUN] and do not execute it.

On completion, emit:

For format: human (default):

══════════════════════════════════════════════
OUTCOME SUMMARY
  Discounts scanned:     <n>
  Expired:               <n>
  Zero usage (> <n> days): <n>
  Total flagged:         <n>
  Deleted:               <n>
  Errors:                <n>
  Output:                discount_cleanup_<date>.csv
══════════════════════════════════════════════

For format: json, emit:

{
  "skill": "discount-hygiene-cleanup",
  "store": "<domain>",
  "started_at": "<ISO8601>",
  "dry_run": true,
  "outcome": {
    "scanned": 0,
    "flagged_expired": 0,
    "flagged_zero_usage": 0,
    "deleted": 0,
    "errors": 0,
    "output_file": "discount_cleanup_<date>.csv"
  }
}

Output Format

CSV file discount_cleanup_.csv with columns:

discount_id, title, status, created_at, ends_at, usage_count, usage_limit, flag_reason, action

Error Handling

Best Practices