Catalog Services MCP Server
Enables interaction with Adobe Commerce Catalog Services to retrieve product variants, price overrides, category permissions, and environment details via MCP.
README
Catalog Services MCP Server
An MCP (Model Context Protocol) server for Adobe Commerce Catalog Services, providing tools to interact with product, category, and environment services via gRPC and REST APIs.
Proto Schemas from https://git.corp.adobe.com/magento/proto-schemas
Features
This MCP server provides the following tools:
- Product Variants: Get product variant information (configurable product options)
- Product Overrides: Query product price overrides by customer group and website
- Category Permissions: Retrieve category permission information
- Environment Details: Get environment, website, and store view information
Prerequisites
- Python 3.11 or higher
- uv - Fast Python package installer and resolver
- Cursor IDE (or any MCP-compatible client)
Installation
1. Install uv
If you don't have uv installed, install it first:
# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Or using pip
pip install uv
2. Clone and Setup the Project
# Clone the repository
git clone <repository-url>
cd catalog-services-mcp
# Install dependencies using uv
uv sync
3. Configure Environment Variables
Set the required environment variables for authentication:
# Adobe Commerce Catalog Service credentials
export CATALOG_SERVICE_API_KEY="your_api_key"
export CATALOG_SERVICE_PRIVATE_KEY="your_private_key"
# New Relic credentials (if using New Relic tools)
export NEW_RELIC_API_KEY="your_new_relic_api_key"
4. Configure MCP in Cursor
Add the following configuration to your Cursor MCP settings file (~/.cursor/mcp.json):
{
"mcpServers": {
"Catalog Services": {
"command": "uv",
"args": [
"--directory",
"/Users/YOUR_USERNAME/path/to/catalog-services-mcp",
"run",
"main.py"
]
}
}
}
Important: Replace /Users/YOUR_USERNAME/path/to/catalog-services-mcp with the actual path to your cloned repository.
5. Restart Cursor
After adding the configuration, restart Cursor IDE to load the MCP server.
Usage
Once configured, you can use the Catalog Services tools in Cursor by invoking them through the MCP protocol. Available tools include:
get_product_variants- Retrieve product variant informationget_product_overrides_by_website- Get product price overrides for all customer groupsget_category_permissions- Query category permissionsget_environment_details- Get environment informationget_environment_store_views- List all store views for an environmentfind_store_view_codes- Find store views by website/store code
Development
Project Structure
catalog-services-mcp/
├── catalog_service/ # Main service module
│ ├── generated/ # Generated protobuf files
│ ├── proto-schemas/ # Proto schema definitions
│ ├── product_service_tool.py
│ ├── product_override_service_tool.py
│ ├── variant_service_tool.py
│ ├── category_permission_service_tool.py
│ └── environment_service_tool.py
├── main.py # MCP server entry point
├── pyproject.toml # Project dependencies
└── README.md
Regenerating Protobuf Files
If you need to regenerate the protobuf files from the proto schemas:
cd catalog_service
./generate_protos.sh
Troubleshooting
MCP Server Not Starting
-
Check that
uvis installed and in your PATH:which uv -
Verify the path in
mcp.jsonis correct and absolute -
Check Cursor logs for error messages
Authentication Errors
Ensure your environment variables are set correctly:
echo $CATALOG_SERVICE_API_KEY
echo $CATALOG_SERVICE_PRIVATE_KEY
Connection Issues
- Verify network connectivity to Adobe Commerce services
- Check that you're connected to the corporate VPN if required
- Ensure firewall rules allow outbound connections
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Submit a pull request
License
Internal Adobe Commerce project
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。