Quicken

v1.0.0Finance & Fintechstable

A python based mcp server (with embedded duckdb) to interact with Quicken data via an LLM Client.

mcpmcp-serverpersonal-financequicken
Share:
8
Stars
0
Downloads
0
Weekly
0/5

What is Quicken?

Quicken is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to python based mcp server (with embedded duckdb) to interact with quicken data via an llm client.

A python based mcp server (with embedded duckdb) to interact with Quicken data via an LLM Client.

This server falls under the Finance & Fintech category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • A python based mcp server (with embedded duckdb) to interact

Use Cases

Query personal finance data from Quicken using embedded DuckDB.
Analyze spending and budgets through AI.
sgoley

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedApr 28, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx quicken

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Quicken

The Quicken MCP Server converts your exported Quicken QIF (Quicken Interchange Format) file into a queryable DuckDB in-memory database and exposes it through MCP tools. You can list accounts, filter transactions, run arbitrary SELECT queries, and generate financial summaries — all from Claude or any MCP-compatible AI client, without exposing your data to the internet.

Prerequisites

  • Docker installed (recommended) or Python 3.10+ with pip for local development
  • A Quicken QIF export file from your Quicken installation (File > Export > QIF)
  • Claude Desktop or another MCP client that supports stdio transport
  • Git to clone the repository if running locally without Docker
1

Export your Quicken data as a QIF file

In Quicken, use File > Export > QIF File to export your accounts and transactions. Save the resulting .qif file to a known location such as ~/data/personal-data.qif. This file is the only input the MCP server needs.

2

Build the Docker image

Clone the repository and build the Docker image. Docker is the recommended approach because it sandboxes the server and avoids Python dependency conflicts. The image is small and builds quickly.

git clone https://github.com/sgoley/quicken-mcp-server.git
cd quicken-mcp-server
docker build -t quicken-mcp-server .
3

Test the server manually

Run the Docker container directly to confirm it loads your QIF file and starts the MCP stdio server. You should see log output indicating the number of accounts and transactions imported into DuckDB.

docker run --rm \
  -v "$PWD/data:/data:ro" \
  quicken-mcp-server \
  --qif /data/personal-data.qif
4

Add the server to Claude Desktop

Edit claude_desktop_config.json to register the Docker-based server. Replace the path with the absolute path to the directory containing your QIF file. The -v flag mounts it read-only into the container for safety.

5

Verify tools are available in Claude

After restarting Claude Desktop, ask Claude to list your accounts. If the tools appear and return data, the server is working correctly. You can also run the MCP Inspector to debug tool calls interactively.

# Test with MCP Inspector (SSE mode)
docker run --rm \
  -v "$PWD/data:/data:ro" \
  -p 8700:8700 \
  quicken-mcp-server \
  --qif /data/personal-data.qif \
  --server-mode sse \
  --listen 0.0.0.0:8700

npx @modelcontextprotocol/inspector http://127.0.0.1:8700/sse

Quicken Examples

Client configuration

Claude Desktop configuration using Docker to run the Quicken MCP server. The QIF file directory is mounted read-only so the server can never modify your financial data.

{
  "mcpServers": {
    "quicken-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-v",
        "/absolute/path/to/data:/data:ro",
        "quicken-mcp-server",
        "--qif",
        "/data/personal-data.qif"
      ],
      "env": {}
    }
  }
}

Prompts to try

Natural language prompts that exercise the Quicken MCP tools for account overview, transaction analysis, and financial reporting.

- "List all my accounts and their current balances"
- "Show me all grocery transactions from last month"
- "What were my top 10 spending categories in 2024?"
- "Run a SQL query to find all transactions over $500 in the checking account"
- "Generate a monthly summary of income versus expenses for Q1 2024"
- "Search for any transactions mentioning 'Amazon'"

Troubleshooting Quicken

Docker container exits immediately with a QIF parse error

Ensure the QIF file was exported correctly from Quicken. Open the file in a text editor and verify it starts with '!Account' or '!Type:'. Some Quicken versions export in different formats — try exporting a single account first to confirm the format is valid QIF.

Claude Desktop shows no tools from the quicken-mcp server

Check that Docker is running before starting Claude Desktop, since the server process is launched on demand. Also confirm the -v path in the args array uses the correct absolute path to your data directory and that the QIF filename matches exactly.

run_sql tool returns 'only SELECT statements are allowed'

This is expected security behavior — the server restricts all queries to SELECT only. Rewrite any INSERT, UPDATE, or DELETE queries as SELECT queries. Your source QIF file is also mounted read-only, so no modifications to the original data are possible.

Frequently Asked Questions about Quicken

What is Quicken?

Quicken is a Model Context Protocol (MCP) server that python based mcp server (with embedded duckdb) to interact with quicken data via an llm client. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Quicken?

Follow the installation instructions on the Quicken GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Quicken?

Quicken works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Quicken free to use?

Yes, Quicken is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Finance & Fintech MCP Servers

Explore all finance & fintech servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "quicken": { "command": "npx", "args": ["-y", "quicken"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Quicken?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides