Note from the author - This project came about due to the sudden popularity and prowess of the DeepSeek LLM family. I have always been fascinated by web development and wanted to see if I could learn Electron by building a code editor which suits my workflow. This project was a great experience to learn about open source and local LLMs. Building NarwhAI taught me not just about Electron and Monaco Editor, but about the incredible ecosystem of open-source AI tools democratizing access to artificial intelligence.

NarwhAI - Local AI Code Editor

A privacy-focused code editor with local AI assistance powered by Ollama and Monaco Editor

Main editor interface with AI panel and multi-tab support

Features

• 100% Local & Private - Your code never leaves your machine
• AI-Powered Assistance - Generate, debug, and explain code using local LLMs
• Multi-Tab Interface - Work with multiple files simultaneously
• Simplistic Yet Functional UI - Clean, modern UI with beige/aqua/blue color scheme
• Auto-Starts Ollama - No manual terminal commands needed
• Smart File Management - Proper save/save-as functionality with visual indicators
• Monaco Editor - Industry-standard code editor (same as VS Code)

Quick Start

Prerequisites

1. Node.js (v16 or higher) - Download
2. Ollama - Download

Installation

# Clone the repository
git clone https://github.com/yourusername/narwhai.git
cd narwhai

# Install dependencies
npm install

# Start the application
npm start

First Run Setup

On first launch, NarwhAI will automatically start Ollama. You'll need to install at least one model:

# Recommended models (choose one or more)
ollama pull deepseek-coder:6.7b  # Best for coding
ollama pull llama3.2:3b          # Lightweight & fast
ollama pull deepseek-r1:7b        # Advanced reasoning

Keyboard Shortcuts

Action	macOS	Windows/Linux
New Tab	⌘ T	Ctrl+T
Close Tab	⌘ W	Ctrl+W
Open File	⌘ O	Ctrl+O
Save	⌘ S	Ctrl+S
Save As	⌘ ⇧ S	Ctrl+Shift+S
Generate Code	⌘ G	Ctrl+G
Debug Code	⌘ D	Ctrl+D
Explain Selection	⌘ E	Ctrl+E

Features in Detail

AI Capabilities

• Generate Code: Create code from scratch or enhance existing code
• Debug Analysis: Identify bugs and get improvement suggestions
• Explain Selection: Get detailed explanations of selected code snippets

Tab System

• Visual indicators for unsaved changes (• dot)
• Confirmation prompts for closing unsaved files
• Each tab maintains its own file path and state
• Automatic language detection based on file extension

Supported Languages

JavaScript, TypeScript, Python, HTML, CSS, JSON, Markdown, YAML, and more - full Monaco Editor language support.

Configuration

Adding More Models

Edit index.html to add more Ollama models to the dropdown:

<select id="model-select">
    <option value="deepseek-coder:6.7b">DeepSeek Coder</option>
    <option value="llama3.2:3b">Llama 3.2</option>
    <option value="deepseek-r1:7b">DeepSeek R1</option>
    <!-- Add your models here -->
    <option value="codellama:7b">CodeLlama</option>
</select>

Building for Distribution

macOS

npm run build-mac
# Creates .dmg in dist/ folder

Windows

npm run build-win
# Creates installer in dist/ folder

Project Structure

narwhai/
├── main.js              # Electron main process
├── preload.js           # Secure IPC bridge
├── renderer.js          # UI and Monaco integration
├── index.html           # Application interface
├── package.json         # Dependencies & configuration
├── build/
│   └── icons/          # Application icons
│       ├── icon.png    # Linux/dev icon
│       ├── icon.icns   # macOS icon
│       └── icon.ico    # Windows icon
└── node_modules/        # Dependencies (git-ignored)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Monaco Editor - The powerful code editor
• Electron - Cross-platform desktop framework
• Ollama - Local LLM inference engine
• DeepSeek - For the excellent coding models
• Inspired by Claude's elegant design philosophy

Disclaimer

NarwhAI is an independent project and is not affiliated with Ollama, Microsoft (Monaco), DeepSeek, or Anthropic. This is a community-driven, open-source tool that interfaces with these technologies.

Known Issues

• Ollama must be installed separately before first use
• Some models require significant RAM (8GB+ recommended)
• Auto-start of Ollama may fail if not in system PATH

Support

For issues, questions, or suggestions, please open an issue on GitHub.