QuickBooks Time MCP Server (Combined)

What is QuickBooks Time MCP Server (Combined)?

A combined MCP server that provides unified access to the QuickBooks Time API through a single interface. It merges four separate servers (JobCode Tools, Reports & Core Tools, Timesheet Tools, and User Tools) and is intended for developers and users who need to interact programmatically with QuickBooks Time data.

How to use QuickBooks Time MCP Server (Combined)?

Install dependencies with pip install -r requirements.txt, create a .env file with your QB_TIME_ACCESS_TOKEN, then run python main.py. The server listens for JSON-RPC requests on stdin/stdout and can be configured with Claude Desktop using the provided example settings.

Key features of QuickBooks Time MCP Server (Combined)

• Retrieve jobcodes with advanced filtering and hierarchy
• Fetch timesheets, including currently active ones
• Access user, group, and project management data
• Generate payroll, project, and current totals reports
• Query custom fields, notifications, and managed clients
• Paginated results and date‑range filtering on most endpoints

Use cases of QuickBooks Time MCP Server (Combined)

• Pulling real‑time timesheet data for payroll processing
• Generating project reports with time entries grouped by jobcode
• Auditing user and group information from QuickBooks Time
• Building custom dashboards that combine jobcode hierarchy and totals
• Automating notifications and custom field data extraction

FAQ from QuickBooks Time MCP Server (Combined)

What authentication is required?

A valid QuickBooks Time access token must be provided via the QB_TIME_ACCESS_TOKEN environment variable.

What are the runtime dependencies?

Python and the packages listed in requirements.txt (installable via pip). The server communicates over stdin/stdout using JSON-RPC.

Can I run it without Claude Desktop?

Yes. The server starts with python main.py and can be integrated with any MCP‑compatible client.

What file structure is expected?

Place main.py in the directory specified in the Claude Desktop configuration (e.g., ./qb-time-mcp-server/main.py). The .env file should be in the working directory.

Are there any known limitations?

The project was developed with AI assistance, so contributions and bug reports are especially welcome. All endpoints support pagination with a maximum of 200 results per page.