Deploy Your Own AI Platform with Dux AI: A Step‑by‑Step Guide

Dux AI is an open‑source AI application platform built on Dux PHP Admin that unifies agents, bots, knowledge bases, and workflow into a single backend, offering modular extensions, multimedia capabilities, and integration with business systems, with detailed installation and deployment instructions for developers.

Open Source Tech Hub
Open Source Tech Hub
Open Source Tech Hub
Deploy Your Own AI Platform with Dux AI: A Step‑by‑Step Guide

Overview

Dux AI is an AI application platform built on the Dux PHP Admin framework. It unifies intelligent agents, chatbots, knowledge bases, and workflow management within a single backend architecture, allowing AI capabilities to be deployed, scaled, and integrated into business systems.

Project Positioning

A unified intelligent‑agent platform

A robot integration platform

A knowledge‑base and file‑parsing platform

A workflow and asynchronous‑task platform

An extensible traditional API‑driven backend

Because it is built on Dux PHP Admin, it provides:

Modular extension capability

Backend pages and resource APIs

Built‑in menu, permission, storage, and scheduling infrastructure

Ability to continue extending existing business systems

Core Features

Agents, bots, knowledge bases, and workflows unified in one platform

Agent capabilities are continuously extensible, supporting HTTP, functions, knowledge bases, MCP, multimedia, etc.

Both synchronous and asynchronous tasks are handled, suitable for image/video processing, polling, and complex automation

File parsing, image generation, video generation and other multimedia functions

Robot integration with DingTalk, Feishu, QQ bots, Enterprise WeChat and similar platforms

Designed to work alongside CRM, OA, appointment, e‑commerce, customer‑service, hospital and other business systems

Suitable for traditional API, multi‑client front‑ends and backend development

Target Teams

Small teams that want to build their own AI assistant

Enterprise teams looking to integrate AI into existing business systems

Product teams developing knowledge‑base Q&A, bot services, or workflow automation

Development teams that need to plug AI capabilities directly into existing PHP backends

Quick Start

Environment Requirements

PHP 8.4 or higher

MySQL 8.0+ (recommended)

Composer

Install Project

composer create-project duxweb/dux-ai
cd dux-ai

Start Local Service

Default PHP Built‑in Server

php -S localhost:8000 -t public

FrankenPHP Worker Mode

frankenphp run
# Or load a custom configuration file
frankenphp run --config Caddyfile

Docker Container (or any containerized environment)

php -S 0.0.0.0:8488 -t public

After starting the server, open http://127.0.0.1:8000 (or the port you configured) in a browser.

Access Installation Wizard

http://localhost:8000/

When the wizard finishes, the backend management UI is available at http://localhost:8000/manage/.

Production Deployment Tips

Set the web root directory to the public folder.

Configure standard PHP rewrite rules (e.g., RewriteEngine On and route all requests to index.php).

In the backend, navigate to System → Plan Management and add the AI Scheduler task.

Run the scheduler as a daemon: php dux scheduler:run For a complete deployment guide, refer to the “System Deployment” chapter in the official documentation.

Documentation Structure

Usage : configuration of services, models, agents, robots, knowledge bases, and workflows.

Development : service calls, capability extensions, driver registration, and workflow development.

API : external chat interface and backend management API.

Related Projects

Dux AI Docs: https://github.com/duxweb/dux-ai-docs Dux PHP Admin: https://github.com/duxweb/dux-php-admin Dux Lite: https://lite2.docs.dux.plus/ DVHA:

https://dvha.docs.dux.plus/

Demo Site

Official demonstration:

https://ai.docs.dux.plus
AIdeploymentPHPChatbotKnowledgeBase
Open Source Tech Hub
Written by

Open Source Tech Hub

Sharing cutting-edge internet technologies and practical AI resources.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.