Does iDBQuery connect to my database with read-only access?
Yes — iDBQuery is designed for read-only analytics and works best with a read-only database user, so it can query your data but cannot change it. You supply the credentials and can scope them to read access only, giving you a hard guarantee at the database level that nothing gets modified.
iDBQuery's job is to answer questions, not to change your systems, and its access model reflects that.
- Bring your own credentials. You connect a source using a login you provision, and the recommended practice is a read-only user with `SELECT`-only rights.
- Database-enforced safety. When the account itself is read-only, the guarantee lives in your database, not just in the app — even in principle, iDBQuery cannot insert, update, or delete.
- Scope it down. You can further limit that user to specific schemas, tables, or a read replica, so iDBQuery only ever touches what you intend.
This is reassuring for teams pointing an AI tool at production data for the first time: the blast radius is bounded by the permissions you grant. It also pairs naturally with query-in-place — iDBQuery reads your live data where it lives without copying it out — and with deployment options like your own VPC, the Desktop app, or an air-gapped install for the strictest environments. Every answer still shows its SQL and cites the rows it read, all of it strictly read activity.
Updated 2026-06-22
Related questions
- How does iDBQuery make sure it never writes to or deletes my data?
- Does iDBQuery respect my own database login and its permissions?
- Can I hide or mask specific columns from certain users in iDBQuery?
- How do I connect a Postgres read replica to iDBQuery without exposing it publicly?
- How secure is iDBQuery?