DBX: Lightweight Open-Source Database Client Connects 100+ Databases with AI Assistance
DBX is a lightweight (~25MB), open-source database client supporting 100+ databases, featuring AI-assisted SQL editing, ER diagrams, schema comparison, Redis/MongoDB browsers, plugin system, and three deployment modes: desktop, Docker web interface, and MCP integration for AI coding assistants.
DBX is a cross-platform database connection tool built with Rust (core), Tauri (shell), and Vue (UI). Because it bundles no Java runtime and no embedded browser, the installer stays around 25 MB. It runs on Windows, macOS, Linux, and can also be deployed via Docker as a web application.
Supported Data Sources
The official README lists over 100 databases and services. Relational databases include MySQL, PostgreSQL, SQLite, SQL Server, Oracle, and Chinese domestic systems such as Dameng, OceanBase, TiDB, RenDa JinCang, GaussDB, and Doris. NoSQL stores cover Redis and MongoDB. Messaging and service-discovery systems include Kafka, RocketMQ, RabbitMQ, MQTT, Nacos, and etcd. All appear in a single connection tree, eliminating the need for multiple dedicated clients.
Migration from Existing Tools
Connections can be imported directly from Navicat or DBeaver, so users with dozens of saved endpoints do not have to re-enter them manually.
User Interface Walkthrough
The main window splits into three panes: a connection tree on the left, a SQL editor in the center, and an optional AI panel on the right. The editor provides schema-aware autocomplete for tables and columns. Selected statements can be executed independently without running the entire script.
The AI panel accepts natural-language prompts (e.g., "insert a few Three Kingdoms characters") and generates the corresponding SQL for review before execution. Destructive statements such as DROP TABLE or TRUNCATE trigger an extra confirmation dialog. Supported model backends include Claude, OpenAI, local Ollama instances, and any OpenAI-compatible API endpoint, allowing fully offline operation when schema cannot leave the network.
Data Grid and Export
Result sets support filtering, sorting, and inline cell editing. Before committing edits, the tool displays the generated UPDATE statements for verification. Export formats include CSV, JSON, Excel, and INSERT scripts for bulk loading into another database.
Schema Management and Diagnostics
ER diagrams for visualizing relationships.
Structure comparison between two environments to pinpoint missing columns or type mismatches without manual DDL diffing.
Execution plan viewer for query optimization.
Specialized Browsers
Redis: search by key pattern; inspect strings, hashes, lists, and other structures.
MongoDB: document browsing with support for Atlas-style connection strings.
Plugin Ecosystem
A built-in plugin store offers extensions for S3, Kubernetes, LDAP, and more. Plugins are signed and run in a sandboxed iframe. Developers can create custom connectors using the provided Go or TypeScript SDKs.
Three Deployment Modes
1. Desktop Installation
Windows (WinGet): winget install t8y2.dbx Windows (Scoop):
scoop bucket add dbx https://github.com/t8y2/scoop-bucket
scoop install dbxmacOS (Homebrew): brew install --cask dbx After launch, create a connection by entering host, port, and credentials. The editor works like any SQL client; for example:
SELECT city, COUNT(*) AS order_cnt
FROM orders
WHERE created_at >= CURRENT_DATE - INTERVAL '7 day'
GROUP BY city
ORDER BY order_cnt DESC;2. Docker Web Version for Team Access
Run the container:
docker run -d --pull=always --name dbx -p 4224:4224 \
-v dbx-data:/app/data \
t8y2/dbx:latestAccess the UI at http://<server-ip>:4224. The named volume dbx-data persists connection configs and encryption keys; the container can be replaced without data loss. For faster pulls in China, substitute the image with docker.cnb.cool/dbxio.com/dbx:latest as noted in the official README.
3. MCP Integration for AI Coding Assistants
DBX exposes an MCP (Model Context Protocol) server. In the desktop app, enable Settings → MCP , select allowed connections, and choose permission level (read-only, read-write, or full). Then add the following to Cursor's MCP configuration:
{
"mcpServers": {
"dbx": {
"command": "npx",
"args": ["-y", "@dbx-app/mcp-server"]
}
}
}If DBX runs in Docker and the assistant runs locally, point the MCP server at the web instance and supply the login password:
{
"mcpServers": {
"dbx": {
"command": "npx",
"args": ["-y", "@dbx-app/mcp-server"],
"env": {
"DBX_WEB_URL": "http://localhost:4224",
"DBX_WEB_PASSWORD": "your-login-password"
}
}
}
}Command-Line Interface
For scripting, install the official CLI:
npm install -g @dbx-app/cli
dbx connections list --json
dbx query local "select 1" --jsonReplace local with the connection name saved in DBX.
Source code and releases: https://github.com/t8y2/dbx (Apache-2.0 license).
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
