intuit/quickbooks-online-mcp-server

What it solves

This project provides a bridge between AI models (like Claude Code) and QuickBooks Online, allowing AI agents to interact with financial data. It eliminates the need for manual data entry or manual report generation by giving LLMs the ability to read, create, and manage accounting entities and financial reports directly via the Model Context Protocol (MCP).

How it works

It operates as a local MCP server that runs as a subprocess on a user's machine. It connects to the QuickBooks Online API using OAuth 2.0 authentication. The server exposes 145 distinct tools to the AI client, which the AI can call to perform specific accounting tasks. It includes built-in support for token rotation to maintain secure, long-term access to the company's financial data.

Who it’s for

Developers and business owners who use QuickBooks Online and want to automate their accounting workflows using AI agents or MCP-compatible LLM clients.

Highlights

• Comprehensive API Coverage: Provides 145 tools covering 29 different entity types (such as Customers, Invoices, and Bills) and 11 financial reports (including Balance Sheets and Profit & Loss).
• Full CRUD Capabilities: Supports creating, reading, updating, and deleting data for most entities.
• Granular Access Control: Users can disable specific categories of tools (Write, Update, or Delete) via environment variables to ensure the AI only has read-only access if desired.
• Secure Authentication: Implements OAuth 2.0 with support for both sandbox and production environments.
• Type Safe: Built with TypeScript and Zod for rigorous validation of data passing between the AI and the API.