Antom MCP Server
A Model Context Protocol (MCP) compatible server that integrates Ant International's Antom payment APIs, enabling AI assistants to handle payment and refund operations seamlessly.
README
Antom MCP Server
A Model Context Protocol (MCP) compatible server that integrates Ant International's Antom payment APIs, enabling AI assistants to handle payment and refund operations seamlessly.
Overview
The Antom MCP Server wraps Ant International's Antom payment APIs into standardized MCP tools, allowing AI assistants to securely process payment-related operations during conversations. With this server, you can create payment sessions, query transaction status, handle refunds, and more directly through AI interactions.
Features
💳 Payment Operations
- Create Payment Session (
create_payment_session): Generate payment sessions for client-side SDK integration - Query Payment Details (
query_payment_detail): Retrieve transaction status and information for submitted payment requests - Cancel Payment (
cancel_payment): Cancel payments when results are not returned within expected timeframes
💰 Refund Operations
- Create Refund (
create_refund): Initiate full or partial refunds against successful payments - Query Refund Details (
query_refund_detail): Check refund status for previously submitted refund requests
Prerequisites
Before using the Antom MCP Server, ensure you have:
- Python 3.11 or higher
- uv (recommended package manager) or pip
- Valid Antom Merchant Account with:
- Merchant Client ID (CLIENT_ID)
- Merchant RSA Private Key (MERCHANT_PRIVATE_KEY)
- Alipay RSA Public Key (ALIPAY_PUBLIC_KEY)
- Payment Redirect Return URL (PAYMENT_REDIRECT_URL)
- Payment Notification Callback URL (PAYMENT_NOTIFY_URL)
Quick Start
1. Installation
Direct Usage with uvx (Recommended)
uvx ant-intl-antom-mcp
Install from Source
git clone https://github.com/alipay/global-antom-mcp.git
cd global-antom-mcp
uv install
2. MCP Client Configuration
Add the following configuration to your MCP client:
{
"mcpServers": {
"antom-mcp-server": {
"command": "uvx",
"args": ["ant-intl-antom-mcp"],
"env": {
"GATEWAY_URL": "https://open-sea-global.alipay.com",
"CLIENT_ID": "your_client_id_here",
"MERCHANT_PRIVATE_KEY": "your_merchant_private_key_here",
"ALIPAY_PUBLIC_KEY": "your_alipay_public_key_here",
"PAYMENT_REDIRECT_URL": "/",
"PAYMENT_NOTIFY_URL": "https://your-domain.com/payment/notify"
}
}
}
}
3. Environment Variables
| Variable | Required | Description |
|---|---|---|
GATEWAY_URL |
❌ | Antom API gateway URL (defaults to https://open-sea-global.alipay.com) |
CLIENT_ID |
✅ | Merchant client ID for identity verification |
MERCHANT_PRIVATE_KEY |
✅ | Merchant RSA private key for request signing |
ALIPAY_PUBLIC_KEY |
✅ | Alipay RSA public key for response verification |
PAYMENT_REDIRECT_URL |
❌ | The user is redirected to after the payment is completed |
PAYMENT_NOTIFY_URL |
❌ | Payment result notification callback URL |
Integration Example
Here's how you can integrate the Antom MCP Server with your AI agent (using QwenAgent as an example):
import os
from qwen_agent.agents import Assistant
# Configure the MCP server as a tool
tools = [{
"mcpServers": {
"antom-mcp-server": {
"command": "uvx",
"args": ["ant-intl-antom-mcp"],
"env": {
"CLIENT_ID": os.getenv('CLIENT_ID'),
"MERCHANT_PRIVATE_KEY": os.getenv('MERCHANT_PRIVATE_KEY'),
"ALIPAY_PUBLIC_KEY": os.getenv('ALIPAY_PUBLIC_KEY'),
"GATEWAY_URL": "https://open-sea-global.alipay.com",
"PAYMENT_REDIRECT_URL": "/",
"PAYMENT_NOTIFY_URL": "https://your-domain.com/notify"
}
}
}
}]
# Create your AI assistant with payment capabilities
bot = Assistant(
llm={'model': 'qwen-max', 'api_key': 'your-api-key'},
function_list=tools,
system_message="You are a helpful assistant with payment processing capabilities."
)
Changelog
See CHANGELOG.md for a detailed history of changes.
License
This project is licensed under the MIT License.
Acknowledgments
- Model Context Protocol for the standard
- Antom Integration for the Antom payment platform
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。