Tabularis

Version: 0.9.9

What is Tabularis?

Tabularis is a lightweight, developer-focused database management tool built with Tauri, Rust, and React. It provides a fast, native desktop experience for connecting, exploring, querying, and managing databases — with no cloud, no sign-up, and no telemetry. Extensible through a plugin system.

Key Features

🔌 Multi-Database Support

Native drivers for PostgreSQL (multi-schema), MySQL / MariaDB (multi-database), and SQLite
Manage multiple connection profiles with secure local storage
Passwords and API keys stored in the OS keychain (Keychain Access, Windows Credential Manager, libsecret)
Read-only mode to protect production databases at the application layer
Connection groups for organising profiles in the sidebar

🔒 SSH Tunneling

Full SSH tunneling implementation in Rust with two automatic backends: russh (password auth) and system SSH (key-based, honours ~/.ssh/config)
Dynamic ephemeral port assignment — no manual port configuration
Reusable SSH profiles shared across multiple database connections
Supports ProxyJump / multi-hop chains via ~/.ssh/config

✍️ SQL Editor

Monaco-based editor with syntax highlighting, autocomplete, and multiple tabs
DataGrip-style execution: run selected text, run all, or run single statement
Saved queries panel for frequently used snippets
Each tab maintains its own independent connection context

🎨 Visual Query Builder

Drag-and-drop table canvas — connect columns to define JOINs visually
Add filters, sorting, limits, and aggregate functions (COUNT, SUM, AVG…) without writing SQL
Live SQL preview updates as you build
Export the generated query directly to the SQL Editor

📊 Data Grid

High-performance virtualised grid for large result sets
Inline cell editing, row insertion and deletion
Multi-row copy to clipboard; one-click export to CSV or JSON
Smart read-only mode for aggregated queries

🗄️ Schema Management

Inline editing of table and column properties directly from the sidebar
GUI wizards for creating tables, modifying columns, and managing indexes and foreign keys
Interactive ER Diagram with auto-layout — visualise relationships across the whole schema

🤖 AI Assistant (Experimental)

Generate SQL from natural language and get plain-English explanations of complex queries
Supports OpenAI, Anthropic (Claude), OpenRouter, and Ollama for fully local inference
Context-aware: sends only the relevant schema to the AI, never raw data

🔌 MCP Server

Built-in Model Context Protocol server — expose database schemas and run queries directly from Claude, Cursor, Windsurf, or any MCP-compatible agent
One-click setup wizard generates ready-to-paste config for Claude Desktop and Claude Code

📦 SQL Dump & Import

Export full or schema-level database dumps to .sql with table selection
Re-import .sql files with real-time progress tracking and cancellation support
Streaming write — works on large databases without memory pressure

🪟 Split View

Open two or more connections side-by-side in resizable panes
Each pane has its own SQL editor, data grid, and independent connection state
Useful for comparing environments, migrating data, or cross-database work

📈 Task Manager

Monitor Tabularis and plugin processes in real time
Per-process CPU, RAM, and disk usage; child process tree inspection
Force-kill or restart any plugin without leaving the app

🎨 Themes

10+ built-in themes including Dracula, Nord, Monokai, Solarized, and One Dark Pro
Syntax highlighting is auto-generated from the active UI theme
Switch themes instantly without restarting

🌍 Internationalisation

Available in English, Italian, and Spanish
Automatic language detection with manual override in Settings

🔄 Seamless Updates

Startup check against GitHub Releases API with one-click install
Package-manager-aware: shows a reminder for AUR, Snap, Homebrew, and winget installs instead of the built-in updater

🧩 Plugin System

Install and manage third-party database drivers without rebuilding the app
Plugin registry with one-click install; plugins run in isolated processes
Built-in Task Manager for monitoring plugin resource usage

Supported Databases

Database	Support
MySQL / MariaDB	Full (multi-database)
PostgreSQL	Full (multi-schema)
SQLite	Full
Additional drivers	Via plugins (DuckDB, Redis, ClickHouse, and more)

Available On

macOS — Universal Binary (Intel + Apple Silicon); Homebrew cask available
Windows — 64-bit installer (requires WebView2)
Linux — AppImage, .deb, .rpm, Snap, AUR (tabularis-bin)

tabularis

About this repository