PacketBreaker Docs

Everything you need to get started with PacketBreaker — from installation to writing your first prompt to shipping production code.

Introduction

PacketBreaker is an AI-powered code builder that transforms natural language prompts into production-ready software. It understands your intent, plans the architecture, generates clean code, and helps you debug — all from a single interface.

Whether you're building a full-stack web app, a CLI tool, or a machine learning pipeline, PacketBreaker handles the heavy lifting so you can focus on what matters: shipping.

Installation

Desktop App

Download the installer for your platform:

# Windows
curl -fsSL https://packetbreaker.dev/install.ps1 | powershell

# macOS
brew install packetbreaker

# Linux
curl -fsSL https://packetbreaker.dev/install.sh | bash

PacketBreaker Web

No installation required. Open app.packetbreaker.dev in your browser and start coding instantly. The web version includes full AI-powered editing, real-time suggestions, and cloud project storage.

Quick Start

After installation, launch PacketBreaker and create your first project:

# Initialize a new project
packetbreaker init my-project

# Open the interactive prompt
packetbreaker prompt

# Or pass a prompt directly
packetbreaker build "Create a REST API with Express and MongoDB"

PacketBreaker will analyze your prompt, plan the file structure, generate code, and set up your project — all in seconds.

System Requirements

Writing Prompts

The quality of PacketBreaker's output depends on the clarity of your prompts. Here are some best practices:

• Be specific — "Build a todo app with React, Tailwind, and local storage persistence" is better than "Build a todo app"
• Mention tech stack — PacketBreaker supports 40+ languages and frameworks. Specify your preferences.
• Describe behavior — "When a user clicks submit, validate the form and show inline errors" gives clear direction.
• Iterate — Start with a base prompt, review the output, and refine with follow-up prompts.

Context Engine

PacketBreaker's context engine reads your entire project — file structure, dependencies, existing code patterns — to generate code that fits seamlessly. It understands:

• Import graphs and module relationships
• Naming conventions and code style
• Framework-specific patterns (Next.js, Django, etc.)
• Test file associations
• Configuration files (tsconfig.json, .eslintrc, etc.)

Code Generation

Generated code follows industry best practices by default: proper error handling, type safety (when applicable), clean separation of concerns, and comprehensive comments.

// Example: Generated Express route handler
router.post('/api/users', async (req, res) => {
  try {
    const { name, email } = req.body;
    
    if (!name || !email) {
      return res.status(400).json({ error: 'Name and email are required' });
    }

    const user = await User.create({ name, email });
    res.status(201).json(user);
  } catch (err) {
    console.error('Failed to create user:', err);
    res.status(500).json({ error: 'Internal server error' });
  }
});

Smart Debugging

Paste an error message or describe a bug, and PacketBreaker will analyze the issue, trace the root cause, and suggest a fix. It can also auto-apply patches with your approval.

# Debug a specific error
packetbreaker debug "TypeError: Cannot read property 'map' of undefined"

# Auto-fix mode
packetbreaker fix --auto

Config File

PacketBreaker uses a packetbreaker.config.json file in your project root:

{
  "model": "gpt-4o",
  "context_depth": "full",
  "auto_commit": false,
  "output_style": "clean",
  "language_preference": "typescript",
  "max_tokens": 8192
}

API Keys

Configure your API key via environment variable or the config file:

# Environment variable
export PACKETBREAKER_API_KEY=pk_live_xxxxxxxxxxxx

# Or via CLI
packetbreaker auth login

Model Selection

CLI Commands

Keyboard Shortcuts

Supported Languages

PacketBreaker supports 40+ languages and frameworks including:

JavaScript TypeScript Python Rust Go C++ C# Java Ruby PHP Swift Kotlin Dart Lua Zig Haskell SQL HTML/CSS Shell YAML

Frameworks: React Next.js Vue Svelte Express Django Flask FastAPI Laravel Rails Spring Boot .NET