Agent context
Toast models
4 SQL models that turn the raw Toast 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
- 11Raw tables
- 11Staging
- 6Core
- 4Analytics
Staging cleans each raw table once, core holds the logic, and analytics is the table dashboards and agents read.
- Labor by jobToast's Labor Summary by job: hours and wage cost per business day.grain: business date x restaurant x job · 2 raw tables
- Labor vs salesLabor cost as a share of net sales, and sales per labor hour, per restaurant and business day.grain: business date x restaurant · 10 raw tables
- Menu item salesToast's Product Mix: quantity, gross and net sales and voids per menu item.grain: business date x restaurant x menu item x sales category · 4 raw tables
- Sales summaryToast's Sales Summary: orders, guests, gross and net sales, discounts, tax, service charges, tips and refunds.grain: business date x restaurant x revenue center x dining option · 8 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 Toast's own reports first. Each model's notes say what the synced data can and can't support.
Connect Toast.
Your agents start with context.
Sync Toast to your warehouse, deploy these models, and query them from Claude, Cursor or ChatGPT over MCP. Free for 14 days, no credit card.