# Editor basics SnoutData's editor is a multi-tab workspace with schema-aware assistance. It is built around SQL, and on a MongoDB connection the same editor also writes [native aggregation pipelines](https://docs.snoutdata.com/databases/mongodb#write-a-native-pipeline). ## Tabs and files | Action | Shortcut | | --- | --- | | New tab | Ctrl/Cmd+N | | Open `.sql` file | Ctrl/Cmd+O | | Save | Ctrl/Cmd+S | | Close tab | Ctrl/Cmd+W | | Run | Ctrl/Cmd+Enter | | Command palette | Ctrl/Cmd+Shift+P | ## Schema-aware completion and hover As you type, the editor suggests tables and columns from the connected database, along with SQL keywords. Hover a name to see its type and details. This is driven by a **live schema index** that refreshes as your database changes, so suggestions stay accurate. For predictive, whole-clause suggestions on top of this, see [AI auto-completion](https://docs.snoutdata.com/ai-assistant/auto-completion). ## Workspaces A **workspace** is a folder of `.sql` scripts on disk (similar to DBeaver). SnoutData can list, create, rename, and open scripts from it, and search across them by file name and content. Saving an untitled tab can auto-name it into the current workspace. ## Command palette Press **Ctrl/Cmd+Shift+P** to run any command by name, including switching connections, opening files, and toggling panels.