Webwright: The Ghost in Your Shell 👻💻

Webwright is more than just a terminal shell—it's a transformative tool that gets things done, plain and simple.

In the early days of computing, the command line was king. Webwright is for the tinkerers and hackers who remember that raw power. It taps into that primal, user-first ethos, acting as a ghost in your shell: an AI that speaks your language and understands your needs.

With Webwright, you're in control. Generate code, manage projects, deploy apps, and automate tasks—all from your terminal. It's the tool that puts power back in your hands.

This is computing at its rawest, most powerful, and most liberating.

Are you ready to unleash the ghost in your shell?

🔑 API Requirements

Webwright requires an API token from either OpenAI or Anthropic to function. You can obtain these tokens from:

• OpenAI API Keys
• Anthropic API Keys

Please ensure you have at least one of these API keys before proceeding with the installation.

Ollama Support

Webwright now supports both local and remote Ollama instances:

Local Ollama

1. Install Ollama following instructions at https://ollama.ai
2. Start the Ollama service
3. Webwright will automatically detect and use the local Ollama instance

Remote Ollama (substrate.run)

1. Get your substrate.run endpoint URL
2. Configure Webwright with your endpoint:

   export OLLAMA_API_ENDPOINT="https://your-instance.substrate.run"

🚀 Key Features

• 🌐 AI-Powered Web Development: Craft and launch websites with intelligent, AI-driven tools.
• 💻 Smart Code Generation: Let AI write code for you, boosting productivity and innovation.
• 📊 Effortless Project Management: Seamlessly create and oversee projects with AI assistance.
• 🔄 Integrated Version Control: Push your code to GitHub without leaving the terminal.
• 🐳 Docker at Your Fingertips: Effortlessly spin up and manage Docker containers.
• 🌐 Browser Magic: Automate web tasks and open URLs (or other apps) with simple commands.
• 🔧 Infinitely Extensible: Customize your shell with bespoke commands and scripts.

🛠️ Installation

Webwright requires Anaconda and Docker to be configured on your system.

1. Install Webwright

   pip install webwright

2. Set up dependencies

Webwright requires Anaconda and Docker to be configured on your system.

• Anaconda/Miniconda Installation
• Docker Desktop Installation

3. Create and activate a Conda environment

   conda create -n webwright python=3.8
   conda activate webwright

4. Install Git Ensure Git is installed in your Conda environment:

   conda install git

5. Start Webwright

   webwright

Getting Started

Once installed, you can start using Webwright by simply typing webwright in your terminal. Here's a quick overview of some commands:

Open URLs in Your Browser

open hackernews

Create a New Project

create project my-project

Generate Code

generate code --type python --output my_script.py

Commit to GitHub

git commit -m "Initial commit"

Start Docker Containers

docker start my-container

AI-Powered Code Generation

Webwright can generate complex code snippets using AI. For example, to generate a fractal:

generate fractal --size 20

Example: Fractal Generation

Here's an example of a Python code snippet generated by Webwright to create a mandlebrot fractal:

import matplotlib.pyplot as plt
import numpy as np

# Function to compute the Mandelbrot set
def mandelbrot(c, max_iter):
    z = c
    for n in range(max_iter):
        if abs(z) > 2:
            return n
        z = z*z + c
    return max_iter

# Generate the fractal
def generate_fractal(size):
    # Determine the plot boundaries
    x_min, x_max = -2.5, 1.5
    y_min, y_max = -2.0, 2.0

    width, height = (size*100, size*100)  # Increase resolution by multiplying size by 100
    x, y = np.linspace(x_min, x_max, width), np.linspace(y_min, y_max, height)
    fractal = np.zeros((width, height))

    for i in range(width):
        for j in range(height):
            fractal[i, j] = mandelbrot(complex(x[i], y[j]), 256)

    plt.imshow(fractal.T, extent=[x_min, x_max, y_min, y_max], cmap='hot')
    plt.colorbar()
    plt.title("Mandelbrot Fractal")
    plt.show()

# Generate a fractal of the given size
generate_fractal(20)

Output

Developer Installation

For developers who want to install Webwright for testing, building, and running from source:

1. Clone the Repository

   git clone https://github.com/MittaAI/webwright.git
   cd webwright

2. Set up Conda Environment

   conda create -n webwright python=3.10
   conda activate webwright

3. Install Dependencies

   pip install -r requirements.txt --no-deps -U chroma-hnswlib==0.7.3

4. Install Git (if not already installed)

   conda install git

5. Add GITHUB_TOKEN to Environment

   $env:GITHUB_TOKEN = "your_github_token"

6. Run Webwright from Source

   python webwright/main.py

Conda Initialization in VSCode

If you are using VSCode and need to get conda working in the editor, you might need to run the following command from the condabin directory:

./conda init powershell

This will set up the necessary configuration for conda to work properly within VSCode terminal sessions.

Building and Testing

To build and test Webwright:

1. Install Development Dependencies Install all dependencies except chroma-hnswlib (current version has issues)

pip install -r <(grep -v chroma-hnswlib requirements.txt)

Install the specific version of chroma-hnswlib

pip install --no-deps chroma-hnswlib==0.7.3

Install remaining dependencies

pip install -r requirements.txt

2. Run Tests

   pytest

NOT WORKING.

3. Build Package

   python setup.py sdist bdist_wheel

Git Setup for Developers

If you're planning to contribute to Webwright, make sure to set up your Git environment:

1. Configure Git

   git config --global user.name "Your Name"
   git config --global user.email "[email protected]"

2. Create a New Branch for Your Changes

   git checkout -b feature/your-feature-name

3. Make Your Changes and Commit

   git add .
   git commit -m "Description of your changes"

4. Push Your Changes and Create a Pull Request

   git push origin feature/your-feature-name

   Then, go to the GitHub repository and create a pull request for your branch.

Remember to always pull the latest changes from the main branch before starting your work:

git checkout main
git pull origin main

For more detailed contribution guidelines, please refer to our CONTRIBUTING.md file.

Documentation

For detailed usage instructions and examples, visit the Webwright Documentation.

Contributing

Webwright is an open-source project. We welcome contributions!

Community and Support

Join our community on Slack for support, discussions, and to share your ideas and feedback.

License

Webwright is open-source software licensed under the MIT License.

© Mitta Corp. All rights reserved 2024.