FullStackGPT

An AI-assisted web development tool that leverages Large Language Models (LLMs) to generate, modify, and fix code for web applications.

Features

• Create Projects: Generate complete web applications from simple descriptions
• Add Features: Enhance existing projects with new functionality
• Fix Bugs: Automatically debug and fix issues in your code
• Run Projects: Launch your web applications locally for testing
• Deploy Projects: Package your applications for deployment
• Multi-Model Support: Compatible with OpenAI, Anthropic, and Google AI models

Installation

1. Clone this repository:

   git clone https://github.com/AidanMercer/fullstackgpt.git
   cd fullstackgpt
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Set up your configuration:

   python main.py setup
   

4. Edit the generated config.yaml file with your API keys:

   llm:
     default_provider: openai  # or anthropic, google
     openai:
       api_key: your_openai_api_key_here
       model: gpt-4-turbo-preview
     anthropic:
       api_key: your_anthropic_api_key_here
       model: claude-3-7-sonnet-20250219
     google:
       api_key: your_google_api_key_here
       model: gemini-pro

Usage

Creating a New Project

Generate a new web application from a description:

python main.py create "a blog platform with user authentication" --project-name my-blog

Adding Features

Add new functionality to an existing project:

python main.py add "add a comment system to blog posts" --project-name my-blog

Fixing Bugs

Automatically fix issues in your code:

python main.py fix "login form doesn't validate email addresses" --project-name my-blog

Running Projects

Launch your application locally:

python main.py run --project-name my-blog

Deploying Projects

Package your application for deployment:

python main.py deploy --project-name my-blog

Listing Available LLM Providers

View configured LLM providers:

python main.py list-providers

Switching LLM Providers

You can specify a different LLM provider for any command:

python main.py create "a task management app" --project-name my-tasks --provider anthropic

Project Structure

Generated projects are stored in the projects/ directory and include:

• Complete Node.js/Express backend
• Frontend HTML, CSS, and JavaScript files
• Package.json with all dependencies
• README.md with setup and usage instructions
• Git repository for version control

Requirements

• Python 3.8+
• Node.js and npm (for running generated projects)
• Git (for version control of projects)
• API keys for at least one LLM provider

Supported LLM Providers

• OpenAI: GPT-4 and other models
• Anthropic: Claude models
• Google: Gemini models

License

MIT License

Contributing

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