SQL Sage for DBeaver

SQL Sage for DBeaver is for teams that work across many database engines from a single client. DBeaver connects to dozens of databases, and SQL Sage adds a dialect-aware AI assistant on top (a DBeaver AI plugin) — without leaving the editor you already use.

Which databases does the DBeaver AI plugin support?

SQL Sage has dedicated, dialect-aware support for the five most common engines: PostgreSQL, MySQL/MariaDB, Oracle, SQL Server and SQLite. It knows their paging syntax, identifier quoting and how each one exposes a query plan, so its suggestions match the database you are actually on — one AI SQL assistant for your whole database stack. Read more about AI for DBeaver: one copilot, five engines. You install the plugin in DBeaver on Windows — and Windows only: Linux and macOS are not supported.

How do I connect my own AI model in DBeaver?

Like the SSMS edition, SQL Sage for DBeaver runs keyless — you use your own Claude, ChatGPT or Codex account, with no API key and no second AI bill. You pay only for the model you already use, and the assistant works right inside the SQL editor, on the connection you have open. Sign-in details are on Claude and ChatGPT in DBeaver, no API key.

Is it safe across many databases at once?

It is read-only by default; writes and DDL run only after you confirm them. Where the engine allows it, a read-only transaction enforces that at the server as well — the same safety discipline holds whether you are on PostgreSQL, Oracle or SQL Server.

What gets sent to the AI model?

The model receives your schema and SQL, not your data — true to the “prove, don’t guess” principle. Rather than guessing, SQL Sage reads real values, table sizes and active sessions straight from the database. It is the choice for anyone who needs one consistent assistant no matter which database they are writing against today — free to try for 30 days.