Getting started

DataTray is a native desktop app — no account, no server to set up. This is the quickest path from install to running SQL against your own database, in about a minute.

Already installed? Skip to Connections for the detail on every field, or to the Query editor.

1. Install DataTray

Grab the build for your OS from the download page — Windows, macOS or Linux. There is no account and no telemetry wall; the installer notes per platform (Gatekeeper, AppImage permissions) are on that page.

Coming from SQL Explorer? DataTray is the same application under a new name, and it brings your setup across by itself — see Upgrading from SQL Explorer below.

2. Add a connection

Click New connection, pick an engine and fill in the details. SQL Server, MySQL / MariaDB, PostgreSQL and SQLite are built in; MongoDB, Redis, DragonflyDB and Elasticsearch install from the Plugin Store. Test first, then Save — credentials go into your OS keychain, never a plain-text file.

See Connections for folders, connection colours and the full field list.

3. Browse your schema

Expand the connection to walk databases, schemas, tables, views and routines. The tree loads lazily, so large servers stay responsive. Double-click a table to open a browse tab — paging, a WHERE filter and column sorting, without writing SQL.

4. Run your first query

Open a SQL tab and start typing, or load a file with Ctrl+O. Press F5 to run. Results land in the grid below, and when a result traces back to a single table with a primary key you can edit it in place — the generated INSERT/UPDATE/DELETE is shown for review and runs in one transaction.

5. Add plugins

Open Tools → Plugin Store to add providers, tools and extensions. The four SQL engines ship with the app; everything else — first-party or third-party — installs the same way, into a per-user folder that needs no admin rights. See Plugin types.

6. Connect an AI assistant (optional)

Enable the MCP server per connection — opt-in and fail-closed — so an AI assistant can query your data over the Model Context Protocol, on a loopback port, with an audit log you control.

Upgrading from SQL Explorer

SQL Explorer became DataTray in 0.6.0. Same application, new name — and installing over an existing SQL Explorer replaces it in place rather than leaving a second copy behind.

On its first start, DataTray copies the old data folder to its own: connections, query history, favorites, keymap, open tabs and installed plugins. Saved passwords come across the first time each connection needs one. You should find everything where you left it.

It copies rather than moves, so an older SQL Explorer build still starts with everything intact. The cost is that the two stop tracking each other from that point on — changes you make in DataTray won't appear in the old build. The old folder is left in place with a MOVED-TO-DATATRAY.txt note pointing at the one that is now live; delete it once you're sure you won't go back.

The folders themselves, if you want to look:

Windows — %APPDATA%\Lionear\DataTray, previously %APPDATA%\Lionear\SqlExplorer
macOS and Linux — ~/.config/Lionear/DataTray, previously ~/.config/Lionear/SqlExplorer

If a DataTray folder already exists, the copy is skipped rather than overwriting it — the newer of the two wins. A copy that fails part-way is rolled back, so the next start retries from a clean slate instead of inheriting half a data folder.

Where to next

Connections →Every field, folders and keychain storage. Query editor →Tabs, .sql files, formatting and history. Keyboard shortcuts →Everything you can rebind. Plugin Store →The engines and tools you can install.