Turn natural language into SQL. Query and design your database with AI.
⬇ Download for macOS · Landing Page · Contributing
Connect to your database, ask questions in plain English, or design schemas through conversation. Get accurate queries and visual ERDs — execute, visualize, and export from a native desktop app.
| 🗣 No SQL expertise required | Describe what you need, get working queries |
| 🔒 Your schema, your data | AI sees your tables and relationships for accurate results |
| 🌐 Works offline | Use local Ollama or connect to cloud AI providers |
| ⚡ One app for everything | Query, chart, and export without leaving your workflow |
| ✏️ Design schemas in plain English | Describe tables and relationships; AI proposes schemas and visual ERDs |
| Feature | Description |
|---|---|
| 🗄️ Multi-database | PostgreSQL, MySQL, SQLite, SQL Server |
| 🤖 AI providers | OpenAI, Anthropic, Google, OpenRouter, or local Ollama |
| 🧠 Schema-aware | AI receives your schema for context-accurate queries |
| ✏️ Schema Designer | AI-powered schema design via chat. Start from scratch or introspect existing DB. Visual ERD canvas, DDL generation, optional execution. Export ERD as PNG/SVG. |
| 🔍 Schema search | Fuzzy search across tables and columns (Cmd+K) |
| 📋 EXPLAIN visualizer | Run EXPLAIN on queries, view plans as Mermaid diagrams |
| ⚡ Query optimization | Get index and rewrite suggestions from AI |
| 📊 Results | Sortable, filterable tables with virtual scrolling |
| 📈 Charts | Bar, line, pie, scatter, area — from any result set |
| 📤 Export | Excel, CSV, and branded Reports (headers, footers, logo, chart) |
| 💬 Conversations | Chat history saved locally |
git clone https://github.com/meetgodhani/sql-assist-desktop.git
cd sql-assist-desktop
npm install
npm run devPrerequisites: Node.js 18+
Databases:
PostgreSQL · MySQL · SQLite · SQL Server
AI Providers:
OpenAI · Anthropic · Google · OpenRouter · Ollama (local)
npm run buildSee CONTRIBUTING.md for macOS code signing, notarization setup, and release workflow details.
src/
├── main/ # Electron main process (services, IPC, database)
├── preload/ # Preload scripts (contextBridge APIs)
├── renderer/ # Vue 3 frontend (components, stores, views)
└── shared/ # Shared types and constants
Contributions are welcome. See CONTRIBUTING.md for setup, code style, and how to submit a pull request.
MIT © meetgodhani