Splitwise MCP Server

A Rust-based Model Context Protocol (MCP) server that provides Claude and other AI agents with access to your Splitwise account for expense tracking and management. This server can be deployed locally or remotely and accessed via HTTP API, making it compatible with web-based AI services.

Features

User Management

• Get current user information
• Retrieve details about other users

Group Management

• List all groups
• Get group details
• Create new groups
• Add/remove users from groups

Expense Management

• List expenses with filters (date range, group, friend)
• Get expense details
• Create new expenses (equal or custom split)
• Update existing expenses
• Delete expenses

Friend Management

• List all friends with balances
• Get friend details
• Add new friends by email

Utilities

• Get supported currencies
• Get expense categories

Prerequisites

• Rust 1.70+ and Cargo
• Splitwise account and API key
• For local: Claude Desktop or other MCP-compatible client
• For remote: Docker and docker-compose (optional)

Installation

1. Clone the repository:

git clone https://github.com/marnunez/splitwise-mcp-server.git
cd splitwise-mcp-server

2. Copy the environment file and add your API key:

cp .env.example .env
# Edit .env and add your SPLITWISE_API_KEY

3. Get your Splitwise API key:

   • Go to https://secure.splitwise.com/apps
   • Register a new application
   • Copy the API key

4. Build the project:

cargo build --release

Configuration

Environment Variables

Create a .env file with:

# Required
SPLITWISE_API_KEY=your_api_key_here

# For HTTP server (remote access)
MCP_AUTH_TOKEN=your_secure_token_here
PORT=8080

# Optional logging level
RUST_LOG=info

Local Setup (Claude Desktop)

Add to your Claude configuration file (claude.json):

{
  "mcpServers": {
    "splitwise": {
      "command": "/path/to/splitwise-mcp-server/target/release/splitwise-mcp",
      "args": [],
      "env": {
        "SPLITWISE_API_KEY": "your_api_key_here",
        "RUST_LOG": "info"
      }
    }
  }
}

Remote Setup (HTTP Server)

Docker Deployment (Recommended)

1. Create .env file with your credentials:

SPLITWISE_API_KEY=your_api_key_here
MCP_AUTH_TOKEN=your_secure_token_here

2. Deploy with docker-compose:

docker-compose up -d

The server will be available at http://your-server:8080/mcp

Manual Deployment

# Build the HTTP server
cargo build --release --bin splitwise-mcp-http

# Run with authentication
SPLITWISE_API_KEY=your_key MCP_AUTH_TOKEN=your_token ./target/release/splitwise-mcp-http

Usage Examples

Once configured, you can ask Claude to:

• View expenses: "Show me my recent Splitwise expenses"
• Create expense: "Add a $50 dinner expense to my 'Roommates' group"
• Check balances: "What are my current balances with friends?"
• Manage groups: "Create a new group called 'Weekend Trip'"
• Track spending: "How much did I spend on groceries last month?"

Available Tools

User Tools

• get_current_user - Get authenticated user info
• get_user - Get user by ID

Group Tools

• list_groups - List all groups
• get_group - Get group details
• create_group - Create new group

Expense Tools

• list_expenses - List expenses with filters
• get_expense - Get expense details
• create_expense - Create new expense
• update_expense - Update expense
• delete_expense - Delete expense

Friend Tools

• list_friends - List friends and balances
• get_friend - Get friend details
• add_friend - Add friend by email

Utility Tools

• get_currencies - List supported currencies
• get_categories - List expense categories

Using with AI Services

ChatGPT (Custom GPT)

1. Create a Custom GPT
2. Add an action with this OpenAPI spec:

openapi: 3.0.0
info:
  title: Splitwise MCP Server
  version: 0.1.0
servers:
  - url

... [View full README on GitHub](https://github.com/marnunez/splitwise-mcp-server#readme)