Skip to main content

GitHub Codespaces Integration

Prerequisites​

  • A GitHub account with access to Codespaces.
  • Access to the daikin-dsv/rad-template repository.
  • If you want to use built-in AI chat, a GitHub plan with GitHub Copilot access.

Setup​

  1. Open daikin-dsv/rad-template.
  2. In the top-right corner, click Use this template -> Open in a codespace.
  3. Wait for Codespaces to initialize dependencies.
  4. Run:
# Use Node v24
nvm use 24
# Copy env example to local file
cp .env.example .env.local
# Build web component bundle
npm run build
# Run dev server
npm run dev

Preview App​

While the dev server is running, open the Ports tab at the bottom to find the forwarded URL:

UI Preview available in forwarded address

Changes are reflected automatically in the forwarded address. The idle timeout is short, so expect the environment to stop after inactivity.

Using AI to Build UI​

In the bottom-right of Codespaces, choose the LLM model available to your account (options vary by subscription/plan).

Use chat to build UI

Ask the model to build the UI you want. The template provides guidance to AI tools via AGENTS.md and CLAUDE.md.

⚠️ Note: Some models may rely on specific files. OpenAI, Claude, and Grok have been tested; other models have not.