Agent context
Sona models
3 SQL models that turn the raw Sona 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
- 7Raw tables
- 7Staging
- 3Core
- 3Analytics
Staging cleans each raw table once, core holds the logic, and analytics is the table dashboards and agents read.
- Absence dailyWho was off, and why, per day: absence days by type and location, with sickness and holiday split out. The basis for absence rate and sickness trend reporting.grain: one row per absence date x home org unit x absence type x day status · 5 raw tables
- Paid hours dailyPaid hours per day as Sona sends them to payroll, by location and pay type: the hours side of a labour cost report.grain: one row per local pay date x home org unit x pay type x entry type · 3 raw tables
- Shift coverage dailyRota coverage per location and shift type: how many shifts were planned, how many were filled by your own staff, how many by an agency, and how many are still open. Covers upcoming shifts as well as past ones, so it doubles as a look-ahead of open shifts to fill.grain: one row per local shift date x org unit x shift type · 2 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 Sona's own reports first. Each model's notes say what the synced data can and can't support.
Connect Sona.
Your agents start with context.
Sync Sona to your warehouse, deploy these models, and query them from Claude, Cursor or ChatGPT over MCP. Free for 14 days, no credit card.