Agentic Product Jsonld Backfill

Purpose

AI shopping agents read a product's structured data (the fields Shopify themes emit as schema.org/Product JSON-LD) to confirm price, availability, and identity. Missing barcodes (GTIN), SKUs, vendor, or product type leave the listing ambiguous — so the agent skips it or recommends a competitor whose data is complete. This skill finds products/variants with those gaps and backfills them: vendor and product type at the product level, barcode/SKU at the variant level. Fixes the agentiq.report findings product-schema-jsonld, gtin-sku-pdp, and variant-metadata.

Prerequisites

Parameters

All skills accept these universal parameters:

Skill-specific parameters:

Safety

> ⚠️ Step 3 (productUpdate) and Step 4 (productVariantsBulkUpdate) write live product/variant data. Barcodes and SKUs are matched from your barcodes_csv; a wrong mapping mislabels a product's identity to every agent. Always run dry_run: true first and verify the change set CSV. This skill never overwrites a field that already has a value — it only fills blanks.

Workflow Steps

Inputs: first: 250, optional query: "tag:''" or collection filter; fields vendor, productType, variants{ id sku barcode }; paginate until hasNextPage: false.

Expected output: Products/variants with missing target fields.

Inputs: per product { id, vendor?, productType? } (only where blank and a value is supplied).

Expected output: Updated product; collect userErrors.

Inputs: per product productId + variants: [{ id, barcode?, inventoryItem: { sku? } }] for blank variant fields.

Expected output: Updated variants; collect userErrors across batches.

GraphQL Operations

# products:query — validated against api_version 2025-01
query BackfillProducts($first: Int!, $after: String, $query: String) {
  products(first: $first, after: $after, query: $query) {
    edges {
      node {
        id
        title
        vendor
        productType
        variants(first: 100) {
          edges { node { id sku barcode } }
        }
      }
    }
    pageInfo { hasNextPage endCursor }
  }
}

# productUpdate:mutation — validated against api_version 2025-01
mutation BackfillProductFields($input: ProductInput!) {
  productUpdate(input: $input) {
    product { id vendor productType }
    userErrors { field message }
  }
}

# productVariantsBulkUpdate:mutation — validated against api_version 2025-01
mutation BackfillVariantFields($productId: ID!, $variants: [ProductVariantsBulkInput!]!) {
  productVariantsBulkUpdate(productId: $productId, variants: $variants) {
    productVariants { id sku barcode }
    userErrors { field message }
  }
}

Session Tracking

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

On start, emit:

╔══════════════════════════════════════════════╗
║  SKILL: <skill name>                         ║
║  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
  <Metric label>:   <value>
  Errors:           0
  Output:           <filename or "none">
══════════════════════════════════════════════

For format: json, emit:

{
  "skill": "<skill-slug>",
  "store": "<domain>",
  "started_at": "<ISO8601>",
  "completed_at": "<ISO8601>",
  "dry_run": false,
  "steps": [
    {
      "step": 1,
      "operation": "<OperationName>",
      "type": "query",
      "params_summary": "<string>",
      "result_summary": "<string>",
      "skipped": false
    }
  ],
  "outcome": {
    "metric_key": 0,
    "errors": 0,
    "output_file": null
  }
}

Output Format

human: counts of products/variants updated per field + a CSV of every change (product, variant, field, old, new). json: { products_updated, variants_updated, by_field{...}, errors, output_file }.

Error Handling

Best Practices