Devonz

Devonz is an AI-powered development agent that helps you build full-stack applications through natural language conversation. Originally built upon bolt.diy, Devonz focuses on speed, efficiency, and a streamlined development experience and have evolved the project into a powerful AI coding agent.

Section	Description
Key Features	Core capabilities and highlights
Tech Stack	Technologies used in the project
Installation	Getting started guide
Configuration	Environment variables and settings
AI Providers	All 19 supported AI providers
Project Structure	Codebase organization
Available Scripts	Development and build commands
Settings	App settings and features
Contributing	How to contribute

Key Features

AI-Powered Development

Feature	Description
Natural Language Building	Describe what you want to build, and Devonz creates it
Multi-Provider Support	19 AI providers including OpenAI, Anthropic, Google, Groq, Ollama, and more
Model Context Protocol (MCP)	Extend Devonz capabilities with MCP tools
Auto-Fix	Automatic error detection and fixing with terminal error detector

Development Environment

Feature	Description
In-Browser Development	Full development environment powered by WebContainers
Real-time Preview	Instant preview of your applications
Terminal Access	Full terminal access within the browser
Code Editor	Integrated CodeMirror editor with syntax highlighting

Deployment Options

Platform	Description
GitHub	Push directly to GitHub repositories
GitLab	Deploy to GitLab projects
Netlify	One-click deployment to Netlify
Vercel	Deploy to Vercel with ease

Integrations

Integration	Description
Supabase	Database and authentication API routes (requires configuration)
Git	Built-in Git support for version control
Template Gallery	Pre-built templates for popular frameworks

Tech Stack

Category	Technology
Framework	Remix + Vite
Language	TypeScript
Styling	UnoCSS + Tailwind CSS
UI Components	Radix UI, Headless UI
Animation	Framer Motion
AI SDK	Vercel AI SDK
Editor	CodeMirror
Terminal	xterm.js
WebContainers	StackBlitz WebContainer API

Installation

Prerequisites

Requirement	Version
Node.js	18.18.0 or higher
pnpm	Latest (recommended)

Quick Start

Note: The first load can take up to 2 minutes before you see the main chat interface. Please be patient while all dependencies are compiled.

Clone the Repository:

git clone https://github.com/zebbern/Devonz.git
cd Devonz

Install Dependencies:
```
pnpm install
```
Start Development Server:
```
pnpm run dev
```
Open in Browser: Navigate to http://localhost:5173

Configuration

Environment Variables

Create a .env.local file in the project root:

# AI Provider API Keys
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_anthropic_key
GOOGLE_GENERATIVE_AI_API_KEY=your_google_key
GROQ_API_KEY=your_groq_key

# Local Provider URLs
OLLAMA_BASE_URL=http://127.0.0.1:11434
LMSTUDIO_BASE_URL=http://127.0.0.1:1234

# Deployment Integrations (Optional)
GITHUB_ACCESS_TOKEN=your_github_token
NETLIFY_AUTH_TOKEN=your_netlify_token
VERCEL_ACCESS_TOKEN=your_vercel_token

In-App Configuration

Click the Settings icon in the sidebar
Navigate to Providers tab
Configure your preferred AI providers:
- Cloud Providers: OpenAI, Anthropic, Google, Groq, OpenRouter, etc.
- Local Providers: Ollama, LM Studio, OpenAI-compatible endpoints

Supported AI Providers

Cloud Providers

Provider	Models	Features
OpenAI	GPT-4o, GPT-4 Turbo, GPT-3.5	Chat, Vision
Anthropic	Claude 3.5 Sonnet, Claude 3 Opus	Chat, Vision
Google	Gemini Pro, Gemini Ultra	Chat, Vision
Groq	LLaMA 3, Mixtral	Fast inference
OpenRouter	100+ models	Model aggregation
Mistral	Mistral Large, Codestral	Chat, Code
Cohere	Command R+	Chat, RAG
Deepseek	Deepseek Coder	Code generation
Amazon Bedrock	Claude, Titan	Enterprise
Together	Open source models	Chat, Code
Perplexity	Online models	Web search
HuggingFace	Open source models	Community models
xAI	Grok	Chat
GitHub	Copilot models	GitHub AI
Hyperbolic	Various models	Specialized inference
Moonshot	Moonshot models	Chinese LLM

Local Providers

Provider	Description
Ollama	Run open-source models locally with model management
LM Studio	Local model inference with GUI
OpenAI-like	Any OpenAI-compatible API endpoint

Project Structure

bolt.diy/
├── app/
│   ├── components/         # React components
│   │   ├── @settings/      # Settings panel (14 tabs)
│   │   ├── auth/           # Authentication components
│   │   ├── chat/           # Chat interface
│   │   ├── deploy/         # Deployment (GitHub, GitLab, Netlify, Vercel)
│   │   ├── editor/         # Code editor
│   │   ├── git/            # Git integration
│   │   ├── header/         # App header
│   │   ├── sidebar/        # Sidebar navigation
│   │   ├── ui/             # Shared UI components
│   │   └── workbench/      # Development workbench
│   ├── lib/                # Core libraries
│   │   ├── api/            # API utilities
│   │   ├── hooks/          # React hooks
│   │   ├── modules/        # Feature modules (llm with 19 providers)
│   │   ├── services/       # API services
│   │   ├── stores/         # State management (nanostores)
│   │   └── utils/          # Utility functions
│   ├── routes/             # Remix routes (39 API endpoints + pages)
│   ├── styles/             # Global styles
│   └── types/              # TypeScript types
├── docs/                   # Documentation (mkdocs)
├── public/                 # Static assets
├── scripts/                # Build scripts
└── supabase/               # Supabase folder (requires configuration)

Available Scripts

Development

Command	Description
`pnpm run dev`	Start development server
`pnpm run build`	Build for production
`pnpm run start`	Run production build
`pnpm run preview`	Build and preview locally

Testing and Quality

Command	Description
`pnpm test`	Run tests
`pnpm test:watch`	Run tests in watch mode
`pnpm run typecheck`	TypeScript type checking
`pnpm run lint`	ESLint check
`pnpm run lint:fix`	Auto-fix linting issues

Utilities

Command	Description
`pnpm run clean`	Clean build artifacts
`pnpm run prepare`	Set up husky git hooks

Settings and Features

Settings Tabs

Tab	Description
Profile	User profile management
Providers	AI provider configuration (19 providers)
Features	Enable/disable features
MCP	Model Context Protocol tools
GitHub	GitHub integration settings
GitLab	GitLab integration settings
Netlify	Netlify deployment settings
Vercel	Vercel deployment settings
Supabase	Database integration settings
Event Logs	Application logs
Data	Import/export data
Notifications	Notification preferences
Project Memory	Project context storage
Settings	General settings

MCP (Model Context Protocol)

Devonz supports MCP tools for extending AI capabilities:

Feature	Description
Custom MCP Servers	Configure custom MCP servers
Specialized Tools	Add specialized tools for your workflow
External Services	Extend AI reasoning with external services

Updating

Git-based Updates

# Save local changes
git stash

# Pull latest updates
git pull

# Update dependencies
pnpm install

# Restore local changes
git stash pop

Clean Install

# Remove dependencies
rm -rf node_modules pnpm-lock.yaml

# Clear cache
pnpm store prune

# Reinstall
pnpm install

Contributing

We welcome contributions! Here's how to get started:

Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Commit your changes: git commit -m 'Add amazing feature'
Push to the branch: git push origin feature/amazing-feature
Open a Pull Request

Acknowledgments

Credit	Description
bolt.diy	Original project foundation
StackBlitz WebContainers	In-browser development environment
Vercel AI SDK	AI capabilities

Links

Link	URL
Repository	https://github.com/zebbern/Devonz
Original Project	bolt.diy

Build anything with AI. Just describe what you want.

Name		Name	Last commit message	Last commit date
Latest commit History 1,705 Commits
app		app
assets/icons		assets/icons
docs		docs
icons		icons
public		public
scripts		scripts
types		types
.editorconfig		.editorconfig
.env.example		.env.example
.env.production		.env.production
.gitignore		.gitignore
.prettierignore		.prettierignore
.prettierrc		.prettierrc
README.md		README.md
connection_test.txt		connection_test.txt
eslint.config.mjs		eslint.config.mjs
package.json		package.json
pnpm-lock.yaml		pnpm-lock.yaml
pre-start.cjs		pre-start.cjs
tsconfig.json		tsconfig.json
uno.config.ts		uno.config.ts
vite.config.ts		vite.config.ts

Folders and files

Latest commit

History

Repository files navigation

Devonz

Table of Contents

Key Features

AI-Powered Development

Development Environment

Deployment Options

Integrations

Tech Stack

Installation

Prerequisites

Quick Start

Configuration

Environment Variables

In-App Configuration

Supported AI Providers

Cloud Providers

Local Providers

Project Structure

Available Scripts

Development

Testing and Quality

Utilities

Settings and Features

Settings Tabs

MCP (Model Context Protocol)

Updating

Git-based Updates

Clean Install

Contributing

Acknowledgments

Links

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages