Libroo Website

This is the landing page for Libroo — a personal library app for tracking your books, managing collections, and keeping track of what you’ve lent to friends.

You can find the main libroo repository here.

✨ Features

• Responsive and fast landing page
• Light/dark mode with a custom toggle (click & right-click interactions)
• Early-access signup modal with email capture
• Spam protection via Cloudflare Turnstile
• Privacy-friendly analytics using Umami
• Configurable legal pages without repository-bundled legal text

🚀 Tech Stack

• Nuxt 4
• Nuxt UI
• Tailwind CSS
• Nuxt SEO
• Cloudflare Workers
• Cloudflare D1
• Cloudflare Turnstile
• Umami Analytics

📦 Setup

git clone https://github.com/niklhut/libroo-website.git
cd libroo-website
pnpm install
pnpm dev

☁️ Deploy To Cloudflare Workers

This repository deploys via GitHub Actions (not Cloudflare Dashboard Git builds).

1. Create a production D1 database.

pnpm wrangler d1 create libroo-website

1. Add GitHub repository secrets.

• CLOUDFLARE_API_TOKEN
• CLOUDFLARE_ACCOUNT_ID
• CLOUDFLARE_D1_DATABASE_ID (production DB UUID)
• CLOUDFLARE_D1_PREVIEW_DATABASE_ID (optional dedicated preview DB UUID)
• NUXT_UI_PRO_LICENSE
• NUXT_TURNSTILE_SECRET_KEY_PRODUCTION
• NUXT_TURNSTILE_SECRET_KEY_PREVIEW (optional; falls back to production secret)

1. Add GitHub repository variables used by nuxt.config.ts vars injection.

• CLOUDFLARE_ACCOUNT_SUBDOMAIN (Workers subdomain prefix used to build preview URLs, e.g. <subdomain>.workers.dev)
• NUXT_PUBLIC_TURNSTILE_SITE_KEY
• NUXT_SITE_URL
• NUXT_PUBLIC_SCRIPTS_UMAMI_ANALYTICS_SCRIPT_INPUT_SRC
• NUXT_PUBLIC_SCRIPTS_UMAMI_ANALYTICS_WEBSITE_ID (optional)
• NUXT_LEGAL_IMPRINT_REDIRECT_URL or NUXT_LEGAL_IMPRINT_MARKDOWN_URL (optional)
• NUXT_LEGAL_PRIVACY_REDIRECT_URL or NUXT_LEGAL_PRIVACY_MARKDOWN_URL (optional)
• NUXT_LEGAL_TERMS_REDIRECT_URL or NUXT_LEGAL_TERMS_MARKDOWN_URL (optional)
• NUXT_LEGAL_CLIENT_PLACEHOLDERS_URL (optional)

1. Workflows.

• Push to main runs .github/workflows/deploy-production.yml.
• Pull request updates run .github/workflows/deploy-preview.yml.
• Closing a pull request runs .github/workflows/cleanup-preview.yml.

Preview deploy behavior:

• Each PR gets a dedicated Worker name: libroo-preview-pr-<number>.
• Each PR gets a dedicated D1 database: libroo-preview-pr-<number>.
• On PR close, the preview Worker and D1 DB are deleted.

Production deploy behavior:

• Uses libroo-website Worker name.
• Applies migrations against production D1 before deploy.
• Routes custom domain only for production deploys.

For local/manual CLI usage, you can still run:

pnpm db:migrate:remote
pnpm deploy

For local pnpm dev, the app applies Drizzle migrations for the local sqlite fallback on server startup.

Legal Pages

The /imprint, /privacy, and /terms routes are intentionally configurable and do not ship legal copy in this repository.

Each page supports two modes:

• Set NUXT_LEGAL_IMPRINT_REDIRECT_URL, NUXT_LEGAL_PRIVACY_REDIRECT_URL, or NUXT_LEGAL_TERMS_REDIRECT_URL to redirect visitors to an external legal page.
• Set NUXT_LEGAL_IMPRINT_MARKDOWN_URL, NUXT_LEGAL_PRIVACY_MARKDOWN_URL, or NUXT_LEGAL_TERMS_MARKDOWN_URL to fetch and render markdown from an HTTPS URL.

Redirect URLs take precedence over markdown URLs. If neither variable is set, the route renders a neutral unconfigured message.

Markdown URLs can point to any public HTTPS source, for example a public Cloudflare R2 object, S3 object, GitHub raw file, or a small Cloudflare Worker endpoint that reads from KV/R2 and returns markdown. Keeping the markdown outside environment variables avoids Worker variable size limits and keeps deployment-specific legal text out of the open-source repository.

Markdown pages can also contain client-rendered placeholders for contact details that should not be present in the initial server-rendered HTML. Add tokens like {{LEGAL_CONTROLLER_NAME}}, {{LEGAL_CONTROLLER_ADDRESS}}, {{LEGAL_CONTACT_EMAIL}}, or {{LEGAL_CONTACT_PHONE}} to the markdown, then set NUXT_LEGAL_CLIENT_PLACEHOLDERS_URL to a public HTTPS JSON object:

{
  "LEGAL_CONTROLLER_NAME": "Example Name",
  "LEGAL_CONTROLLER_ADDRESS": "Street 1\n12345 City\nCountry",
  "LEGAL_CONTACT_EMAIL": "hello@example.com",
  "LEGAL_CONTACT_PHONE": "+49 000 000000"
}

During SSR those tokens are replaced with neutral fallback text. After the page hydrates in the browser, the app fetches /api/legal/placeholders and renders the configured values. This is crawler friction, not access control: clients that run JavaScript or fetch the JSON/API directly can still read the contact data.

📜 License

This project is licensed under the MIT License. See LICENSE for details.