Wise

Provides access to Wise recipient operations via an MCP gateway, including listing, validating, and creating recipients and sending money.
  • python

2

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "sergeiledvanov-mcp-wise": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--init",
        "-e",
        "WISE_API_TOKEN=your_api_token_here",
        "-e",
        "WISE_IS_SANDBOX=true",
        "mcp-wise:latest"
      ],
      "env": {
        "MODE": "stdio",
        "WISE_API_TOKEN": "your_api_token_here",
        "WISE_IS_SANDBOX": "true"
      }
    }
  }
}

Wise MCP Server provides a gateway to the Wise API, exposing recipient-related actions as MCP resources. It handles authentication, profile selection, and can run in a Docker container for easy integration and testing with the Wise Sandbox API.

How to use

You will run an MCP client that talks to the Wise MCP Server to list recipients, validate recipient requirements, create recipients, and send money. Start by choosing whether you want a local (stdio) server or a remote HTTP endpoint if you have one configured. The Docker image lets you run the server as a container and pass your Wise API token and sandbox flag through environment variables. Use the available resources to perform common Wise recipient operations without dealing with raw API calls.

Typical usage patterns include listing recipients for a chosen profile type (personal or business) and optional currency filtering, validating account details before creating a recipient, creating a recipient with the required account details, and sending money to a recipient with an optional payment reference and source of funds.

How to install

Prerequisites you need before installing:

  • Python 3.12 or higher if you install directly
  • uv package manager if you install directly
  • Docker if you use the Docker image
  • Wise API token for authentication

Option 1: Direct installation

# Step 1: Clone the project
git clone https://github.com/sergeiledvanov/mcp-wise
cd wise-mcp

# Step 2: Set up environment
cp .env.example .env
# Edit .env to add your Wise API token
# Step 3: Install dependencies with uv
uv venv
uv pip install -e .

Option 2: Using Docker

If you prefer Docker, build the image and run it with your token and sandbox mode enabled.

To build the image and run it with your token, use the following commands.

# Build the Docker image
docker build -t mcp-wise .

MCP server configuration for Docker

Configure the MCP server to run inside Docker by providing your Wise API token and sandbox flag through environment variables.

{
  "mcpServers": {
    "mcp-wise": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--init",
        "-e", "WISE_API_TOKEN=your_api_token_here",
        "-e", "WISE_IS_SANDBOX=true",
        "mcp-wise:latest"
      ],
      "env": [
        {"name": "WISE_API_TOKEN", "value": "your_api_token_here"},
        {"name": "WISE_IS_SANDBOX", "value": "true"}
      ]
    }
  }
}

Additional setup notes

Update your local MCP configuration file to match your chosen mode. For stdio mode (default) copy the stdio template into place. For HTTP mode, copy the HTTP template. Then start your MCP client to connect to the Wise MCP Server.

Available MCP Resources

The server exposes the following resources to manage Wise recipients and money transfers.

list_recipients

Returns a list of all recipients from your Wise account.

  • profile_type: personal or business (default: personal)
  • currency: optional filter by currency code (e.g., EUR, USD)

get_recipient_requirements

Fetches recipient requirements for creating a new recipient and validates provided account details against those requirements.

  • source_currency: e.g., USD
  • target_currency: e.g., EUR
  • source_amount: amount in source currency
  • profile_type: personal or business (default: personal)
  • account: optional account details to validate

create_recipient

Creates a new recipient with the provided account details that conform to Wise API requirements.

  • profile_type: personal or business (default: personal)
  • account: recipient account details including:
    • accountHolderName
    • currency
    • type (e.g., iban)
    • details: country- and currency-specific fields

send_money

Sends money to a recipient using the Wise API.

  • profile_type: personal or business
  • source_currency: e.g., USD
  • source_amount: amount to send
  • recipient_id: recipient identifier
  • payment_reference: optional, defaults to money
  • source_of_funds: optional

Configuration

Configuration is done via environment variables. Set these values in your environment or a .env file that the MCP server loads.

  • WISE_API_TOKEN: Your Wise API token (required)
  • WISE_IS_SANDBOX: true to use Wise Sandbox API (default: false)
  • MODE: MCP Server transport mode, http or stdio (default: stdio)

Development notes

Project structure centers on a Python implementation with a main entry and a modular API client plus MCP resources. You can extend the server by adding new API client methods and new MCP resources, then wiring them into the app.

Security and best practices

Keep your Wise API token secret. Use the Sandbox API for development and testing, and store tokens securely in environment variables or secret management tools when running in production.

Available tools

list_recipients

Lists all recipients from your Wise account, with optional filtering by profile type and currency.

get_recipient_requirements

Retrieves recipient creation requirements and can validate provided account details.

create_recipient

Creates a new recipient with the required account details including currency and type.

send_money

Sends money to a specified recipient using Wise, with optional reference and funds source.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational