Shopify Admin Bundle Availability Check

Purpose

Walks every product flagged as a bundle (either via Shopify's native requiresComponents mechanic or a bundle.components metafield convention), then verifies each component variant has sufficient inventory to back the bundle's quantity ratio. Surfaces bundles that are listed as in-stock on the storefront but cannot actually be fulfilled because one component has run out. Read-only — no mutations.

Prerequisites

• Authenticated Shopify CLI session: shopify store auth --store <domain> --scopes read_products,read_inventory
• API scopes: read_products, read_inventory

Parameters

Safety

ℹ️ Read-only skill — no mutations are executed. Safe to run at any time. The skill reads inventory and metafields only; it never adjusts component quantities or bundle availability.

Workflow Steps

1. OPERATION: products — query Inputs: first: 250, query: "metafield:<namespace>.<key>:* OR product_type:bundle", select requiresSellingPlan, status, metafield(namespace, key), variants, pagination cursor Expected output: Bundle products and their parent variants; paginate until hasNextPage: false

2. For each bundle, parse component list. Native bundles use productVariant.requiresComponents and productVariant.productVariantComponents. Metafield bundles parse JSON value into [{variantId, quantity}].

3. OPERATION: productVariants — query Inputs: Batched IDs of all unique component variants, select inventoryQuantity, inventoryItem { id }, product { title } Expected output: On-hand quantity per component

4. OPERATION: inventoryItems — query Inputs: Batched component inventory item IDs, select tracked, inventoryLevels(first: 25) { quantities } Expected output: Per-location quantity for each component

5. For each bundle, compute max_buildable_units = floor(min over components of (component_on_hand - safety_stock) / required_qty). Flag bundles where max_buildable_units == 0 (broken bundle) or < min_listed_inventory_threshold.

GraphQL Operations

graphqlCopy code
# products:query — validated against api_version 2025-01
query BundleProducts($query: String!, $after: String, $namespace: String!, $key: String!) {
  products(first: 250, after: $after, query: $query) {
    edges {
      node {
        id
        title
        status
        productType
        metafield(namespace: $namespace, key: $key) {
          id
          value
          type
        }
        variants(first: 100) {
          edges {
            node {
              id
              title
              sku
              inventoryQuantity
              requiresComponents
              productVariantComponents(first: 50) {
                edges {
                  node {
                    quantity
                    productVariant {
                      id
                      sku
                      inventoryQuantity
                      product {
                        id
                        title
                      }
                    }
                  }
                }
              }
            }
          }
        }
      }
    }
    pageInfo {
      hasNextPage
      endCursor
    }
  }
}

graphqlCopy code
# productVariants:query — validated against api_version 2025-01
query ComponentVariantStock($ids: [ID!]!) {
  nodes(ids: $ids) {
    ... on ProductVariant {
      id
      sku
      inventoryQuantity
      product {
        id
        title
      }
      inventoryItem {
        id
        tracked
      }
    }
  }
}

graphqlCopy code
# inventoryItems:query — validated against api_version 2025-01
query ComponentInventoryLevels($ids: [ID!]!) {
  nodes(ids: $ids) {
    ... on InventoryItem {
      id
      tracked
      inventoryLevels(first: 25) {
        edges {
          node {
            location {
              id
              name
            }
            quantities(names: ["available", "on_hand", "committed"]) {
              name
              quantity
            }
          }
        }
      }
    }
  }
}

Session Tracking

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

On start, emit:

Copy code
╔══════════════════════════════════════════════╗
║  SKILL: Bundle Availability Check            ║
║  Store: <store domain>                       ║
║  Started: <YYYY-MM-DD HH:MM UTC>             ║
╚══════════════════════════════════════════════╝

After each step, emit:

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

On completion, emit:

For format: human (default):

Copy code
══════════════════════════════════════════════
BUNDLE AVAILABILITY CHECK
  Bundles inspected:       <n>
  Fully buildable:         <n>
  Constrained (low):       <n>
  Broken (cannot build):   <n>

  Top broken bundles:
    "<bundle>"  Bottleneck: "<component>"  Need: <n>  Have: <n>
  Output: bundle_availability_<date>.csv
══════════════════════════════════════════════

For format: json, emit:

jsonCopy code
{
  "skill": "bundle-availability-check",
  "store": "<domain>",
  "bundles_inspected": 0,
  "fully_buildable": 0,
  "constrained": 0,
  "broken": 0,
  "issues": [],
  "output_file": "bundle_availability_<date>.csv"
}

Output Format

CSV file bundle_availability_<YYYY-MM-DD>.csv with columns: bundle_product_id, bundle_title, bundle_variant_sku, max_buildable_units, bottleneck_component_sku, bottleneck_component_title, bottleneck_required_qty, bottleneck_on_hand, status

Error Handling

Best Practices

• Run daily for stores with many bundles; surface broken bundles before customers can buy something you cannot ship.
• Use safety_stock to keep a buffer for non-bundle sales of the same component — bundles share inventory with standalone variants.
• Pair with inventory-adjustment or low-inventory-restock to action a broken bundle into a reorder.
• For native bundles, requiresComponents: true is authoritative — prefer it over metafield conventions when both exist.
• A bundle with max_buildable_units = 0 should also be temporarily unpublished until the bottleneck component is restocked; consider chaining this skill with product-lifecycle-manager.