payu-intrepos/payu-mcp-server

PayU MCP Server

Seamlessly integrate PayU payment services with your AI tools through Model Context Protocol

Overview

PayU MCP Server provides specialized integration that enables AI tools to securely access PayU payment APIs through the Model Context Protocol (MCP). This integration allows AI assistants to create payment links, retrieve transaction details, access invoice information, manage refunds, and analyze settlement data directly, making your payment workflows smarter and more efficient.

What is the Model Context Protocol (MCP)?

The Model Context Protocol (MCP) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. Whether you're building an AI-powered IDE, enhancing a chat interface, or creating custom AI workflows, MCP provides a standardized way to connect LLMs with the context they need.

— Model Context Protocol

Available Tools

Getting Started

Prerequisites

• Python 3.10 or higher
• PayU Merchant Account with API access
• MCP-compatible AI client (Claude AI, etc.)

Installation

Option 1: Standard Installation

1. Clone the Repository

git clone https://github.com/payu-intrepos/payu-mcp-server.git
cd payu-mcp-server

2. Install Dependencies

We recommend using uv for Python dependency management. Install uv from Astral.

curl -LsSf https://astral.sh/uv/install.sh | sh

# Create virtual environment and activate it
uv venv
source .venv/bin/activate

# Install dependencies
uv add "mcp[cli]" httpx python-dotenv

Alternatively, you can use pip:

pip install -r requirements.txt

Option 2: Docker Installation

1. Clone the Repository

git clone https://github.com/payu-intrepos/payu-mcp-server.git
cd payu-mcp-server

2. Build Docker Image

docker build -t payu-mcp-server .

3. Run Docker Container

Run the container with your PayU credentials:

docker run -p 8888:8888 \
  -e CLIENT_ID="YOUR_CLIENT_ID" \
  -e CLIENT_SECRET="YOUR_CLIENT_SECRET" \
  -e AUTH_TOKEN="YOUR_BEARER_TOKEN" \
  -e MERCHANT_ID="YOUR_MERCHANT_ID" \
  -e PORT=8888 \
  payu-mcp-server python server.py --sse --port 8888

3. Add Server to MCP Client Configuration

Example for Claude AI Desktop Client:

Location of configuration file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

For Standard Installation:

Add the following to your configuration:

{
  "mcpServers": {
    "payu-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/payu-mcp-server",
        "run",
        "server.py"
      ],
      "env": {
        "CLIENT_ID": "YOUR_CLIENT_ID",
        "CLIENT_SECRET": "YOUR_CLIENT_SECRET",
        "AUTH_TOKEN": "YOUR_BEARER_TOKEN",
        "MERCHANT_ID": "YOUR_MERCHANT_ID"
      }
    }
  }
}

For Docker Installation:

Configure your MCP client to connect to the Docker container:

{
  "mcpServers": {
    "payu-mcp-server": {
      "url": "http://localhost:8888"
    }
  }
}

API Credentials

To obtain your API credentials, visit the PayU Developer Portal.

Required credentials:

• CLIENT_ID: Your PayU client ID
• CLIENT_SECRET: Your PayU client secret
• AUTH_TOKEN: Your PayU bearer token (preferred method)
• MERCHANT_ID: Your PayU merchant identifier

Additional Permission Scopes

When configuring your API access, ensure you have the following permission scopes:

• read_transactions:

  • Provides access to transaction history and details
  • Allows retrieval of transaction metadata.
  • Enables detailed transaction analysis and tracking

• read_invoices:

  • Enables fetching comprehensive invoice details
  • Provides access to invoice status, amounts, and related transaction information
  • Read-only permission ensures secure access to invoice data

• read_refunds:

  • Access to refund transaction data
  • Enables refund status tracking and analytics

• read_settlements:

  • Access to settlement information
  • Enables settlement tracking and reporting

Note: Contact PayU support to confirm these specific scopes are available and properly configured for your merchant account.

Usage Examples

Creating a Payment Link

Once configured, your MCP Client can create payment links with a natural language request:

1. Payment link with contacts

Create a payment link for ₹5000 for Web Development Services and send to ABC

2. Payment link with email

Create a payment link for ₹5000 for Web Development Services and send to <abc@example.com>

Fetching transactions for Invoice Number/Payment Link

Once configured, your AI assistant can help you fetch invoice number details with a natural language request:

Get the invoice details for <Invoice-ID>

Fetching details for any transaction

Once configured, your AI assistant can help you fetch transaction details with a natural language request:

Provide the transaction details for <Transaction-ID>

Get Transactions List

Get transactions list from 2024-01-01 00:00:00 to 2024-01-31 23:59:59 with status captured

Get Transactions Summary

Get transactions summary from 2024-01-01 00:00:00 to 2024-01-31 23:59:59 for UPI payments

Search Refunds

Search refunds from 2024-01-01 to 2024-01-31 with status success

Get Refunds Summary

Get refunds summary from 2024-01-01 to 2024-01-31

Get Settlement Details

Get settlement details for settlement ID <SETTLEMENT_ID>

Advanced Features

Transaction Filtering Options

The enhanced transaction tools support comprehensive filtering:

• Status Filters: captured, failed, pending, cancelled, etc.
• Payment Modes: UPI, CC, DC, NB, EMI, etc.
• Payment Sources: pg, button, paymentLink, apiIntInvoice, etc.
• Amount Ranges: min/max amount filtering
• Currency Filters: USD, AED, AUD, CAD, GBP, OTH
• Date Ranges: Flexible date range queries
• Additional Filters: IVR, mobile, web, chargeback, etc.

Refund Management

• Search refunds by date range and status
• Get refund summaries and analytics
• Support for multiple refund statuses: requested, success, failure, queued, pending, user_cancelled

Settlement Tracking

• Access settlement details by settlement ID
• Support for UTR references and transaction IDs
• Multiple settlement status tracking

License

This project is licensed under the MIT License - see the file for details.