I built debba.sql, a lightweight, cross-platform database manager because I wanted a native, fast tool for quick data edits without the heaviness of Electron-based giants or Java IDEs.
The interesting part? This project started as an experiment in vibe coding. I utilized an AI agent to handle the entire scaffolding and heavy lifting of the implementation. We went from an empty directory to a working app with SSH tunneling, schema introspection, and data editing in record time. It allowed me to focus purely on the architectural decisions, UX "vibe", and solving my specific pain points.
The Tech Stack: * Frontend: React 19 + TypeScript + Tailwind CSS (using Monaco Editor for SQL) * Backend: Rust Tauri v2 (for native performance and small binary size) * Database Access: SQLx (pure Rust, async driver)
Current Features: * Multi-DB Support: PostgreSQL, MySQL/MariaDB, and SQLite. * SSH Tunneling: Built-in support for connecting via bastion hosts (supports key/agent/password auth). * Native Performance: Fast startup (~10MB binary) and low memory footprint. * SQL Editor: Full syntax highlighting and execution. * Data Grid: Inline editing, row deletion, and easy "New Row" insertion. * Schema Tools: Visual "Create Table" wizard (DataGrip style) and introspection. * Export: Native CSV/JSON export.
Why I built it: I often need to jump into a DB via SSH to run a query, edit a specific cell, or check a schema. I wanted something that opens instantly and feels like a native system utility, not a web browser tab.
Roadmap / Help Wanted: It's currently in Alpha (v0.1.0). I'm looking for feedback, bug reports, and contributions! Next on the list: * Multiple Query Tabs * Visual Query Builder * Dark/Light mode toggle
The code is open source. Check it out here: https://github.com/debba/debba.sql
Let me know what you think about the tool or the AI-assisted dev process!