name: shopify-sdk-patterns description: 'Apply production-ready patterns for @shopify/shopify-api including typed GraphQL clients,
session management, and retry logic.
Use when implementing Shopify integrations, refactoring SDK usage,
or establishing team coding standards for Shopify.
Trigger with phrases like "shopify SDK patterns", "shopify best practices",
"shopify code patterns", "idiomatic shopify", "shopify client wrapper".
' allowed-tools: Read, Write, Edit version: 1.0.0 license: MIT author: Jeremy Longshore jeremy@intentsolutions.io tags:
- saas
- ecommerce
- shopify compatibility: Designed for Claude Code
Shopify SDK Patterns
Overview
Production-ready patterns for the @shopify/shopify-api library: singleton clients, typed GraphQL operations, session management, cursor-based pagination, codegen-typed operations, bulk operations, and webhook registry patterns.
Prerequisites
@shopify/shopify-apiv9+ installed- Familiarity with Shopify's GraphQL Admin API
- Understanding of async/await and TypeScript generics
Instructions
Step 1: Typed GraphQL Client Wrapper
Initialize a singleton shopifyApi instance with LATEST_API_VERSION, cache sessions per shop, and expose a typed shopifyQuery<T>() helper that wraps client.request().
See Typed GraphQL Client for the complete implementation.
Step 2: Error Handling with Shopify Error Types
Custom ShopifyServiceError class that distinguishes retryable errors (429, 5xx) from permanent ones. Includes handleShopifyError() for error translation and safeShopifyCall() that returns {data, error} tuples instead of throwing.
See Error Handling for the complete implementation.
Step 3: Cursor-Based Pagination
Async generator paginateShopify<T>() for Relay-style cursor pagination. Yields batches of nodes, automatically following pageInfo.endCursor until hasNextPage is false. Memory-efficient for large datasets.
See Cursor Pagination for the complete implementation.
Step 4: Multi-Tenant Client Factory
ShopifyClientFactory class for apps installed on multiple stores. Creates isolated GraphqlClient instances per merchant with session caching. Includes removeClient() for eviction on app uninstall.
See Multi-Tenant Factory for the complete implementation.
Step 5: Codegen-Typed Operations
Use @shopify/api-codegen-preset to generate TypeScript types from your GraphQL operations. This eliminates manual type definitions and catches schema changes at build time.
// codegen.ts — project root config
import { shopifyApiProject, ApiType } from "@shopify/api-codegen-preset";
export default {
schema: "https://shopify.dev/admin-graphql-direct-proxy",
documents: ["src/**/*.{ts,tsx}"],
projects: {
default: shopifyApiProject({
apiType: ApiType.Admin,
apiVersion: "2025-04", // Update quarterly
outputDir: "./src/types",
}),
},
};
// src/operations/products.ts — typed query with codegen output
import type { ProductsQuery } from "../types/admin.generated";
const PRODUCTS_QUERY = `#graphql
query Products($first: Int!, $after: String) {
products(first: $first, after: $after) {
edges {
node { id title status totalInventory }
}
pageInfo { hasNextPage endCursor }
}
}
` as const;
// Return type is fully inferred from codegen
export async function getProducts(shop: string): Promise<ProductsQuery> {
return shopifyQuery<ProductsQuery>(shop, PRODUCTS_QUERY, { first: 50 });
}
Run npx graphql-codegen after changing any GraphQL operation or upgrading API versions.
Step 6: Bulk Operation Helpers
For datasets too large for pagination (100k+ records), use Shopify's Bulk Operations API. It runs a query server-side and produces a JSONL file you download when ready.
// src/shopify/bulk.ts
const BULK_QUERY = `
mutation bulkOperationRunQuery($query: String!) {
bulkOperationRunQuery(query: $query) {
bulkOperation { id status }
userErrors { field message }
}
}
`;
const POLL_QUERY = `{
currentBulkOperation {
id status errorCode objectCount url
}
}`;
export async function runBulkOperation(
shop: string,
query: string
): Promise<string> {
// Start the bulk operation
const { bulkOperationRunQuery } = await shopifyQuery(shop, BULK_QUERY, { query });
if (bulkOperationRunQuery.userErrors?.length) {
throw new Error(bulkOperationRunQuery.userErrors[0].message);
}
// Poll until complete (typically 1-10 minutes for large datasets)
let result;
do {
await new Promise((r) => setTimeout(r, 5000)); // 5s interval
result = (await shopifyQuery(shop, POLL_QUERY)).currentBulkOperation;
} while (result.status === "RUNNING" || result.status === "CREATED");
if (result.status !== "COMPLETED") {
throw new Error(`Bulk operation failed: ${result.errorCode}`);
}
return result.url; // JSONL download URL
}
// Usage: export all products
const url = await runBulkOperation(shop, `{
products { edges { node { id title status variants { edges { node { sku price } } } } } }
}`);
const response = await fetch(url);
const jsonl = await response.text();
const products = jsonl.trim().split("\n").map(JSON.parse);
Step 7: Webhook Registry Patterns
Programmatically register webhook subscriptions using webhookSubscriptionCreate with typed WebhookSubscriptionInput. Supports both HTTP and EventBridge/PubSub endpoints.
// src/shopify/webhooks.ts
const REGISTER_WEBHOOK = `
mutation webhookSubscriptionCreate(
$topic: WebhookSubscriptionTopic!,
$webhookSubscription: WebhookSubscriptionInput!
) {
webhookSubscriptionCreate(topic: $topic, webhookSubscription: $webhookSubscription) {
webhookSubscription { id topic }
userErrors { field message }
}
}
`;
interface WebhookConfig {
topic: string;
callbackUrl: string;
format?: "JSON" | "XML";
}
export async function registerWebhooks(
shop: string,
webhooks: WebhookConfig[]
): Promise<{ registered: string[]; errors: string[] }> {
const registered: string[] = [];
const errors: string[] = [];
for (const wh of webhooks) {
const result = await shopifyQuery(shop, REGISTER_WEBHOOK, {
topic: wh.topic,
webhookSubscription: {
callbackUrl: wh.callbackUrl,
format: wh.format ?? "JSON",
},
});
const userErrors = result.webhookSubscriptionCreate.userErrors;
if (userErrors?.length) {
errors.push(`${wh.topic}: ${userErrors[0].message}`);
} else {
registered.push(wh.topic);
}
}
return { registered, errors };
}
Output
- Type-safe GraphQL client with singleton session management
- Structured error handling that distinguishes retryable from permanent errors
- Cursor-based pagination generator for large datasets
- Multi-tenant client factory for apps serving multiple stores
- Codegen-typed operations eliminating manual type definitions
- Bulk operation helpers for large dataset exports
- Webhook registry patterns for programmatic subscription management
Error Handling
| Pattern | Use Case | Benefit |
|---|---|---|
safeShopifyCall | All API calls | Returns {data, error} instead of throwing |
handleShopifyError | Error translation | Maps HTTP/GraphQL errors to typed errors |
| Cursor pagination | Large datasets | Memory-efficient streaming with backpressure |
| Bulk operations | 100k+ records | Server-side execution, no client memory pressure |
| Client factory | Multi-tenant apps | Isolated sessions per merchant |
Examples
Setting Up a Type-Safe GraphQL Client
Initialize a singleton Shopify client with session caching and a typed shopifyQuery<T>() helper for all API calls.
See Typed GraphQL Client for the complete implementation.
Handling Retryable vs Permanent Errors
Distinguish 429/5xx retryable errors from permanent validation failures using a structured error class and safe call wrapper.
See Error Handling for the complete error handling implementation.
Building a Multi-Tenant App
Create isolated GraphQL clients per merchant with session caching and eviction on app uninstall using a client factory.
See Multi-Tenant Factory for the complete implementation.