Skip to main content

MCP Server Configuration

This guide covers how to configure the Cube2 MCP server for MetrePay integrations.

Cube2 is an MCP (Model Context Protocol) server built by Palomonte Labs that enables AI tools to understand, generate, and maintain integrations with financial APIs.


MCP Server Path Access​

The Cube2 MCP server path is provided directly by Palomonte Labs.

To request your Cube2 MCP server path, please email:
support@palomonte.ai

You will receive the directory path required to complete the configuration below.

Supported Tools​

Cube2 works with any tool that supports the Model Context Protocol:

ToolConfig FileDocumentation
Claude Desktopclaude_desktop_config.jsonClaude MCP Docs
Cursorcursor.mcp.jsonCursor Docs
Continueconfig.jsonContinue Docs
CustomVariesSee MCP Specification

Cube2 MetrePay Configuration​

Add the following to your MCP configuration file:

{
"mcpServers": {
"cube2_metrepay": {
"command": "/path/to/uv",
"args": [
"--directory",
"/path/to/mcp1/metrepay/cube_2_metrepay",
"run",
"assistant.py"
]
}
}
}

Configuration File Locations​

OSClaude Desktop Config Path
macOS~/Library/Application Support/Claude/claude_desktop_config.json
Windows%APPDATA%\Claude\claude_desktop_config.json
Linux~/.config/Claude/claude_desktop_config.json

Available Cube2 Tools​

Once configured, Cube2 provides these mapping tools:

ToolPurpose
hi_cube2Welcome message with business scenarios
map_attributes_main_process_toolInteractive attribute mapping
get_bank_semantic_schemaGet semantic entity definitions
get_use_cases_fileRead use case XML mappings
get_best_and_useGet API best practices

Testing Your Setup​

After configuration, test by asking:

"hi cube2"

You should see a welcome message with MetrePay payment link scenarios.

Then try a mapping request:

"Map a payment link with amount, customer email and name"

Choose your LLM tool for detailed setup: