A Model Context Protocol (MCP) server that lets you interact with your Lunchmoney transactions through Claude and other AI assistants.
This server provides three main tools:
- get-recent-transactions: View your recent transactions from the past N days
- search-transactions: Search transactions by keyword in payee names or notes
- get-category-spending: Analyze spending in specific categories
You can use this server directly in Claude Desktop without installation:
{
"mcpServers": {
"lunchmoney": {
"command": "npx",
"args": ["-y", "lunchmoney-mcp-server"],
"env": {
"LUNCHMONEY_TOKEN": "your_token_here"
}
}
}
}
Replace your_token_here
with your Lunchmoney API token, which you can get from your Lunchmoney developer settings.
Once configured in Claude Desktop, you can ask questions like:
- "Show me my recent transactions from the past week"
- "Search for all transactions at Amazon"
- "How much did I spend on restaurants last month?"
The Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to Large Language Models (LLMs). Think of MCP like a USB-C port for AI applications - it provides a standardized way to connect AI models to different data sources and tools.
Some key benefits of MCP:
- Standardized way to expose data and functionality to LLMs
- Human-in-the-loop security (all actions require user approval)
- Growing ecosystem of pre-built integrations
- Works with multiple AI models and applications
To develop locally:
- Clone this repository
- Install dependencies:
npm install
- Build the TypeScript code:
npm run build
- Run with your API token:
LUNCHMONEY_TOKEN=your_token_here node build/index.js
MIT
Contributions are welcome! Please feel free to submit a Pull Request.