ecommerce-mcp
MCP server for commerce-operations self-service. Lets an AI agent look up order details and search orders without needing developer or SQL access.
README
ecommerce-mcp
MCP server for commerce-operations self-service. Lets an AI agent look up order details and search orders without needing developer or SQL access.
Quick Start
# Install deps
npm install
# Seed the database
npm run seed
# Build
npm run build
# Run dev
npm run dev
MCP Tools
get_order(orderId)— Look up a single order by ID (e.g.ORD-001). Returns full details including items, tracking, and notes.search_orders(filters)— Search orders by status, customer, email, date range. Optionally limit results.
Development
- TypeScript,
@modelcontextprotocol/sdk,better-sqlite3 - SQLite database at
data/orders.db(committed to repo since it holds only synthetic data) - MCP communicates over stdio (Cloudflare Tunnel wraps the process for remote access)
Deploy to VPS
# On the VPS
git clone https://github.com/ritz541/ecommerce-mcp.git
cd ecommerce-mcp
npm ci --include=dev
npm run build
npm run seed # run once after first deploy, or if data needs reset
npm start # run as a systemd service for production
Cloudflare Tunnel
Add this entry to /etc/cloudflared/config.yml:
ingress:
- hostname: mcp.chavanpatil.com
service: http://localhost:8003
Then reload the tunnel: systemctl reload cloudflared.
You'll also need to set up a systemd service for the MCP server so it starts on boot and restarts on failure. Create /etc/systemd/system/ecommerce-mcp.service:
[Unit]
Description=Commerce Operations MCP Server
After=network.target
[Service]
Type=simple
User=root
WorkingDirectory=/root/Code/Projects/github/ecommerce-mcp
ExecStart=/usr/bin/node dist/index.ts
Restart=always
Environment=NODE_ENV=production
[Install]
WantedBy=multi-user.target
Then:
systemctl daemon-reload
systemctl enable ecommerce-mcp
systemctl start ecommerce-mcp
GitHub Actions CI
On push to master, the workflow runs builds and deploys to the VPS via SSH. Set these repository secrets in your GitHub repo settings:
| Secret Name | Description |
|---|---|
VPS_HOST |
VPS IP address (172.236.176.58) |
VPS_SSH_KEY |
SSH private key (root user) |
What's In / Out of Scope
- In:
get_order,search_orders, synthetic order data, remote hosting - Out: Auth, frontend, inventory management, fulfillment tools, real payments, Docker
License
ISC
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。