by nocobase
NocoBase is an open-source AI + no-code platform for building business systems fast. Instead of generating everything from scratch, AI works on top of production-proven infrastructure and a WYSIWYG no-code interface, so you get both speed and reliability.
# Add to your Claude Code skills
git clone https://github.com/nocobase/nocobaseLast scanned: 5/1/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-01T06:37:28.604Z",
"semgrepRan": false,
"npmAuditRan": false,
"pipAuditRan": true
}English | 简体中文 | 日本語 | Français | Español | Português | Bahasa Indonesia | Tiếng Việt | Deutsch
https://github.com/user-attachments/assets/3b89d965-f60f-48e0-8110-24186c2911d2
NocoBase is an open-source AI + no-code platform for building business systems fast. Instead of generating everything from scratch, AI works on top of production-proven infrastructure and a WYSIWYG no-code interface, so you get both speed and reliability.
Homepage:
https://www.nocobase.com/
Online demo:
https://demo.nocobase.com/new
Documentation:
https://docs.nocobase.com/
Forum:
https://forum.nocobase.com/c/english-forum/5
User stories:
https://www.nocobase.com/en/blog/tags/customer-stories
Our release notes are updated regularly on the blog, with weekly summaries of important changes.
Coding agents get a full CLI and skills, while people get a WYSIWYG no-code interface, so both can collaborate efficiently.
Go from deployment to a working system in minutes with mainstream coding agents.

People can build and modify visually in a WYSIWYG interface, even without AI.

Split the work as needed: people can refine what AI builds, and AI can continue from human configuration.

NocoBase includes AI employees, so AI can work directly inside the system.
AI employees get business context automatically and execute tasks directly inside the system.

MCP, HTTP APIs, CLI, and rich skills let external agents connect securely.

Every AI action follows the same fine-grained permissions as human users.

Data models, permissions, and workflows are complex and error-sensitive.
NocoBase provides them as built-in infrastructure, tested and proven in production.
Dozens of built-in modules cover the most common business needs.

Business data stays in standard relational structures, separate from the UI.

With a microkernel design, everything is a plugin and the system can grow without losing control.

The simplest way to let an AI agent operate NocoBase is to install the NocoBase CLI, finish initialization, and then start or restart your AI agent session inside the initialized working directory.
Minimal flow:
npm install -g @nocobase/cli@beta
mkdir my-nocobase && cd my-nocobase
nb init --ui
Then restart your AI agent session in that directory, for example:
cd my-nocobase && codex
Learn more:
https://docs.nocobase.com/ai/quick-start
NocoBase supports three installation methods:
Install with Docker (recommended)
Best for no-code scenarios and requires no code writing. To upgrade, pull the latest image and restart.
Install with create-nocobase-app
The business code of your project stays independent and supports low-code development.
Install from Git source code
If you want the latest unreleased version or plan to contribute by modifying and debugging source code directly, this method is recommended. It requires stronger development skills, and you can pull updates through Git when the code changes.
No comments yet. Be the first to share your thoughts!