lightrail-jupyterlab

Connect JupyterLab to Lightrail AI


Keywords
code-generation, desktop, electron, gpt-3, gpt-4, llm, openai, react, typescript
License
Other
Install
pip install lightrail-jupyterlab==0.1.1

Documentation

🚈 Lightrail

Lightrail is an open-source platform for building React/Tailwind front-end UIs with the help of LLMs. It aspires to create clean, well-organized codebases that can be easily exported and edited manually. Support for adding interactive functionality within the Lightrail platform is coming soon. It is currently most useful for scaffolding a UI / layout, then exporting it and adding functionality separately.

Demo

Lightrail.Demo.mp4

Try It Out

The hosted version of Lightrail is available at: https://lightrail.ai

Dependencies / Structure

Lightrail is a Next.js application that relies on the OpenAI GPT-3.5/4 API as well as Supabase.

Running / Developing Locally

  1. Clone this repository
  2. Create a file called .env.local in the top-level directory, with these contents:
OPENAI_API_KEY=...your OpenAI API Key....
NEXT_PUBLIC_SUPABASE_URL=...your Supabase instance URL...
NEXT_PUBLIC_SUPABASE_ANON_KEY=...your Supabase anon key...
NEXT_PUBLIC_LOGIN_REDIRECT_URL="http://localhost:3000/auth/callback"
NEXT_PUBLIC_SERVER_URL="http://localhost:3000"
  1. Make sure Auth in Supabase is configured to use the specified server / redirect URLs
  2. Run npm install
  3. Run npm run dev
  4. Lightrail should now be running locally. Navigate to http://localhost:3000/projects to begin using it.

Roadmap

Lightrail is very much a work-in-progress, and you're likely to run into bugs. Please report any bugs you find in the Issues section of this repository. PRs are also welcome! Below is a rough roadmap of development plans:

  • Generating Complete Component Trees
  • Editing individual components
  • User accounts
  • Exporting to Next.js
  • Creating new components
  • Supporting interactivity (partially done)
  • Uploading images
  • Generating images
  • Linking Data Sources / DBs