Databases & CRM
Set up a SQL database (PostgreSQL, MySQL, SQL Server)
Connect a SQL database — PostgreSQL (including Supabase), MySQL / MariaDB, or Microsoft SQL Server — read with a read-only role. It works out the shape of your data from the schema itself: which tables are your customers, invoices and contracts, and how they relate. Each becomes a record the assistant reasons over alongside your documents — “what do we know about this client”, “how much have they been invoiced” — with the figures totalled exactly from the data. Nothing is ever written.
- Create a READ-ONLY database role (one that can SELECT but not change anything) with access to the schema you want read.
- Supabase: Project → Settings → Database → Connection string — copy it and paste it into the Connection string field.
- Any other database: choose the type (PostgreSQL, MySQL/MariaDB or SQL Server), then fill in host, port, database and a read-only username and password; SSL mode covers PostgreSQL and MySQL/MariaDB (SQL Server encrypts when the server requires it). Amazon Redshift speaks the PostgreSQL dialect — choose PostgreSQL for it.
- Optionally limit to specific tables (or exclude some); blank reads the whole schema.
- Documents → SQL database, enter the connection, choose access groups, Test, then Sync.
Only SELECTs are ever issued, and the session is additionally opened read-only with a query timeout where the engine supports it (PostgreSQL, MySQL — SQL Server has no session read-only mode, so use a read-only role there). Sensitive columns (passwords, card numbers, salaries…) are never ingested — not in records, samples or totals. Large tables aren't copied row by row: each customer/contract/etc. becomes one record with its figures summarised, refreshed on every sync, and very large tables are bounded — the first 5,000 records per table (the sweep summary says so whenever the cap is hit).
Last updated 29 Aug 2026