Shufersal MCP Server
Provides automated shopping capabilities for the Shufersal website using Puppeteer, enabling LLMs to search products, create shopping lists, and add items to shopping carts.
README
Shufersal MCP Server
A Model Context Protocol server that provides automated shopping capabilities for the Shufersal website using Puppeteer. This server enables LLMs to interact with Shufersal's online shopping platform, search for products, create shopping lists, and add items to your cart.
Key Features
- Product Search: Search Shufersal's product catalog, logged in user will get personalized results
- Cart Management: Add products to your Shufersal shopping cart
- Browser Automation: Seamless interaction with the Shufersal website
- Shopping List Management: Create structured shopping lists from web links (e.g. recipe links)
- Console Monitoring: Track console logs from the browser automation (for debugging)
Workflow
- Create Shopping List: Use
create_shopping_list_tableto create a shopping list from your request or recipe link - Open Shufersal: Use
open_shufersalto navigate to the Shufersal website (login required) - Search Products: Use
search_shufersalto find products matching your shopping list items - Add to Cart: Use
add_to_shufersal_cartto add found products to your shopping cart
Components
Tools
-
open_shufersal
- Open the Shufersal website and prepare for shopping (requires user login)
- Input:
hasCreatedShoppingList(boolean): Whether a shopping list has been created beforehand
-
search_shufersal
- Search for products on Shufersal website (results sorted by purchase history)
- Input:
query(string): Product search query (e.g., 'milk', 'bread', 'tomatoes')
-
add_to_shufersal_cart
- Add a product to the shopping cart (must be used after searching)
- Inputs:
product_id(string): Product ID from search resultssellingMethod(string): Selling method from search resultsqty(number): Quantity to add to cartcomment(string, optional): Optional comment for the product
Extra Tools for paste a recipe link
-
read_webpage_content
- Read and convert webpage content to markdown format
- Input:
url(string): URL of the webpage to read
-
create_shopping_list_table
- Create a shopping list table in markdown format from recipe ingredients
- Input:
recipe(array): Array of recipe ingredients with name, quantity, unit of measure, and optional brand
Resources
The server provides access to:
- Console Logs (
console://logs)- Browser console output in text format
- Includes all console messages from the browser automation
Configuration
Here's the Claude Desktop configuration to use the Shufersal MCP server:
{
"mcpServers": {
"shufersal": {
"command": "node",
"args": ["<path/to/shufersal-mcp>/dist/index.js"],
}
}
}
With custrom user data directory:
{
"mcpServers": {
"shufersal": {
"command": "node",
"args": ["<path/to/shufersal-mcp>/dist/index.js", "--user-data-dir=<path/to/custom/user/data>"],
}
}
}
Windows:
{
"mcpServers": {
"shufersal": {
"command": "<path/to/node>",
"args": ["<path/to/shufersal-mcp>/dist/index.js"],
}
}
}
Security Notice
IMPORTANT: This tool automates browser interactions with the Shufersal website and stores browser data locally. Please be aware:
- Browser session data is stored in
./puppeteer-user-data/(excluded from git) - If you log into Shufersal during first use, the tool will save your session data in the user data directory
- Only use this tool with trusted MCP clients
Development
Building
npm run build
Development Mode
npm run watch
Linting
npm run lint
npm run lint:fix
License
This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License.
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。