For Stakeholders

What this is

This PR is the first step of the Watchtower initiative (Jira epic WT-1 / XLZ-556): an internal management and subscription control plane for Dentolize's own team, not a clinic-facing feature. This specific PR does not build Watchtower — it builds and proves the deployment path for Watchtower, using a two-file, disposable scaffold. Full context lives in Confluence ("Watchtower — Master Plan", Docs 81264692).

Why it matters

De-risking the deploy pipeline before the real backend and UI work starts is a small, cheap step: it confirms the sandbox can detect, build, and route to a brand-new package pair (watchtower-server / watchtower-web), the same way it already does for other internal apps like pharmacy. The two engineers who pick up WT-1 next can start on the real product without also debugging infrastructure.

Risk

Low. The PR is explicitly marked do not merge — it's for sandbox review only — and adds no dependencies, no database changes, and no production code paths. The added server has no authentication and returns only static/fixed responses, so there is no data-exposure surface beyond "the scaffold is reachable," which is expected for a sandbox smoke test.

What's not yet decided or built

Everything about the real product: the data model, RBAC design, 2FA flow, audit logging, and UI. None of it is in this PR. Treat any specifics beyond "internal control plane, GraphQL + REST backend, reuses Dashboard/Admin credentials + 2FA" as provisional until the WT-1 epic branches land.