Shopify Storefront MCP Server
Provides access to the Shopify Storefront API via MCP, enabling AI assistants to query and interact with Shopify store data such as products, collections, carts, and customer information.
README
Shopify Storefront MCP Server
This server provides access to the Shopify Storefront API via MCP, allowing AI assistants to query and interact with your Shopify store data.
Features
- Access to product, collection, and inventory data
- Cart creation and management
- Support for GraphQL queries and mutations
- Automatic token handling and validation
- Easy integration with MCP-compatible AI assistants
Setup Instructions
- Clone this repository
- Install dependencies:
pip install -r requirements.txt - Copy
.env.exampleto.envand configure your environment variables - Generate a Storefront API token via Shopify Admin (see below)
- Run the server:
python -m shopify_storefront_mcp_server
Environment Variables
Create a .env file using the provided .env.example as a template:
# Required
SHOPIFY_STOREFRONT_ACCESS_TOKEN=your_storefront_token
SHOPIFY_STORE_NAME=your-store-name
# Optional
SHOPIFY_API_VERSION=2025-04
SHOPIFY_BUYER_IP=127.0.0.1
Generating a Storefront API Token
- Log in to your Shopify admin
- Go to Apps and sales channels > Develop apps > Create an app
- Name your app (e.g., "MCP Storefront")
- Go to API credentials > Configure Storefront API scopes
- Select necessary scopes:
unauthenticated_read_product_listingsunauthenticated_read_product_inventoryunauthenticated_read_product_pricingunauthenticated_write_checkoutsunauthenticated_read_content
- Save and copy the generated Storefront API access token
- Add the token to your
.envfile asSHOPIFY_STOREFRONT_ACCESS_TOKEN
Usage Examples
Running with the MCP server:
python -m shopify_storefront_mcp_server
The server exposes the following MCP tools:
shopify_discover: Detect if a URL belongs to a Shopify storefront and discover authentication tokensshopify_storefront_graphql: Execute GraphQL queries against the Storefront APIcustomer_data: Unified tool for all customer data operations (Create, Read, Update, Delete)
Customer Resources
This server also provides MCP resources for customer information:
customer://name: Customer's full namecustomer://email: Customer's email addresscustomer://phone: Customer's phone numbercustomer://shipping_address: Customer's shipping address (including address1, address2, city, state, postal_code, country)customer://billing_address: Customer's billing address (including address1, address2, city, state, postal_code, country)customer://profile: Complete customer profile
Customer data is stored in user_data/customer.json and should be managed using the customer_data tool.
Managing Customer Data
The server provides a unified customer_data tool for managing all customer information. This tool consolidates create, read, update, and delete operations into a single interface.
Examples:
# Get all customer data
customer_data(operation="get")
# Get a specific field
customer_data(operation="get", field="name")
customer_data(operation="get", field="shipping_address")
# Update a specific field
customer_data(operation="update", field="name", value="Jane Doe")
customer_data(
operation="update",
shipping_address={
"address1": "123 Main St",
"address2": "Apt 4B",
"city": "New York",
"state": "NY",
"postal_code": "10001",
"country": "US"
}
)
# Add custom fields
customer_data(
operation="update",
custom_fields={
"preferences": {
"theme": "dark",
"notifications": "email",
"language": "en-US"
},
"loyalty_tier": "gold",
"last_purchase_date": "2023-06-15"
}
)
# Get a custom field
customer_data(operation="get", field="preferences")
customer_data(operation="get", field="loyalty_tier")
# Update single custom field
customer_data(operation="update", field="loyalty_tier", value="platinum")
# Delete a specific field
customer_data(operation="delete", field="phone")
customer_data(operation="delete", field="preferences")
# Delete all customer data
customer_data(operation="delete")
This consolidated tool simplifies integration with AI assistants by providing a consistent interface for all customer data operations, including both standard customer information and any custom fields that may be useful for personalization.
Data Privacy & Storage
Customer data is stored in user_data/customer.json. This file contains personal information and should not be committed to version control. The repository includes:
user_data/customer.json.example: A template file showing the expected structure with dummy data- Entries in
.gitignoreto prevent accidental commits of actual customer data
When deploying this server, the user_data/customer.json file will be created automatically when the customer_data tool is first used. You can also copy and rename the example file to get started:
cp user_data/customer.json.example user_data/customer.json
All data stored in the customer file persists between server restarts. The file supports both standard customer fields (name, email, addresses) and arbitrary custom fields for AI personalization.
Creating Checkouts with Customer Data
The server makes it easy to create Shopify checkouts that include customer information by combining the customer_data and shopify_storefront_graphql tools.
Example workflow:
# Step 1: Get customer data
customer_profile = customer_data(operation="get")
# Step 2: Create a cart with GraphQL
cart_mutation = """
mutation createCart($lines: [CartLineInput!]!) {
cartCreate(input: {lines: $lines}) {
cart {
id
checkoutUrl
}
userErrors {
field
message
}
}
}
"""
cart_variables = {
"lines": [
{
"merchandiseId": "gid://shopify/ProductVariant/12345678901234",
"quantity": 1
}
]
}
cart_result = shopify_storefront_graphql(
mode="execute",
host="your-store.myshopify.com",
token="your_storefront_token",
query=cart_mutation,
variables=cart_variables
)
# Step 3: Apply customer attributes to the cart
cart_id = # extract from cart_result
customer_info = json.loads(customer_profile)
attributes_mutation = """
mutation updateCartAttributes($cartId: ID!, $attributes: [AttributeInput!]!) {
cartAttributesUpdate(cartId: $cartId, attributes: $attributes) {
cart {
id
checkoutUrl
}
userErrors {
field
message
}
}
}
"""
attributes_variables = {
"cartId": cart_id,
"attributes": [
{
"key": "email",
"value": customer_info["email"]
},
{
"key": "deliveryAddress",
"value": json.dumps(customer_info["shipping_address"])
}
]
}
shopify_storefront_graphql(
mode="execute",
host="your-store.myshopify.com",
token="your_storefront_token",
query=attributes_mutation,
variables=attributes_variables
)
This approach gives you complete control over the checkout process while leveraging the stored customer information.
Troubleshooting
If you encounter authentication errors:
- Verify token format: Storefront API tokens should start with
shpsa_(newer) orshpat_(older) - Check store name: Ensure SHOPIFY_STORE_NAME is correct (without .myshopify.com)
- Check API version: Make sure the API version is supported
- Test token: Use cURL to test your token directly:
curl -X POST \ https://your-store.myshopify.com/api/2025-04/graphql.json \ -H "Content-Type: application/json" \ -H "X-Shopify-Storefront-Access-Token: your_token" \ -d '{"query": "query { shop { name } }"}' - Regenerate token: If issues persist, create a new token with proper scopes
Security Considerations
- Never commit your
.envfile or any files containing API tokens - Use environment variables for all sensitive information
- Consider setting up IP restrictions in your Shopify Admin
- Review the permissions granted to your Storefront API token
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。