client.claritycare.ai

hermine-metrics

Private customer-facing pages (ROI reports, metrics dashboards) served via GitHub Pages at client.claritycare.ai.

The repository is internal/private — the source and history are not publicly browsable. The published Pages site is public-by-link only, and each page lives under a long, unguessable path so it can’t be found by browsing or guessing. Pages also carry a noindex meta tag, so they stay out of search engines.

⚠️ This is obscurity, not access control. Anyone with the exact link can open the page, and links travel through browser history, referrer headers, and forwards. For genuinely confidential data, put a real auth gate (e.g. Cloudflare Access) in front of it.

Create a new page for a customer

1. Create a new folder in the repo root with a long random alphanumeric name — this becomes the secret path. Generate one with:

   LC_ALL=C tr -dc 'a-zA-Z0-9' < /dev/urandom | head -c 40; echo
   

2. Add an index.html inside that folder with the page content:

   <that-long-string>/index.html
   

3. Commit and push to main:

   git add -A
   git commit -m "Add page for <customer>"
   git push
   

4. GitHub automatically rebuilds and publishes it (usually within a minute) to:

   https://client.claritycare.ai/<that-long-string>/
   

   That’s the link to share with the customer.

Notes

• Keep the CNAME file in the repo root — it binds the site to client.claritycare.ai. Deleting it breaks the custom domain.
• The site root (client.claritycare.ai/) intentionally returns 404 — pages exist only under their secret paths.
• One folder = one page. Use a fresh random string per customer so links can’t be guessed from one another.
• DNS: client is a CNAME → clarity-care-ai.github.io (managed at the claritycare.ai DNS provider).