Agent context
Quickbooks models
4 SQL models that turn the raw Quickbooks tables into the reports people ask about. Deploy them in Weld and your agent queries the definitions behind every number.
How the set is built
- 6Raw tables
- 6Staging
- 5Core
- 4Analytics
Staging cleans each raw table once, core holds the logic, and analytics is the table dashboards and agents read.
- AR AP agingOpen invoices and open bills, bucketed by days past due as of today. The document-level detail behind QuickBooks' A/R Aging and A/P Aging reports.grain: company x side x document · 2 raw tables
- General ledgerEvery debit and credit QuickBooks Online posted, one row per journal line, with the account's classification attached. This is the base for the trial balance and the P&L, and the table to filter when someone asks "what hit this account".grain: company x transaction id x document number x line position · 3 raw tables
- Profit and loss by monthQuickBooks' Profit and Loss report, by month and account, built from the general ledger.grain: company x month x account, income and expense accounts only, months with postings only · 3 raw tables
- Trial balanceMonth-end trial balance per account, built from the general ledger.grain: company x month x account, from the account's first posting to the current month, so an account with no activity in a month still carries its balance · 4 raw tables
- Weld or dbt
- The same SQL in both dialects. Sync the folder with GitHub Sync, or copy the dbt models into your project.
- Check before you rely on it
- Reconcile a month against Quickbooks's own reports first. Each model's notes say what the synced data can and can't support.
Connect Quickbooks.
Your agents start with context.
Sync Quickbooks to your warehouse, deploy these models, and query them from Claude, Cursor or ChatGPT over MCP. Free for 14 days, no credit card.