Continuity · internal system

Work Registry

One honest answer to 'is this actually still open?', across every session and every machine.

Live No public endpoint Registry

The problem it solves

Long-running work spread over many sessions produces phantom tasks: items carried forward for weeks that were quietly finished on another machine, and real gaps that everyone assumed someone else had closed. Memory is not a ledger.

How it works

Open

work items registered when started

→

Attribute

linked to the session doing the work

→

auto-closed from the commit that finished it

→

Reconcile

nightly sweep against reality

→

Project

one cross-machine view of what is genuinely open

Under the hood

A commit message can close an item directly, so the ledger updates as a side effect of doing the work rather than as a chore afterwards.
Claims about something 'still being pending' are grepped against the record before they are ever repeated.
Wrap-up carry-forwards are re-validated against live state, which kills phantom items at the source.
Available on every machine, so the answer does not depend on which one you are sitting at.

Why it matters

The failure mode of an ambitious system is not doing too little. It is losing track of what it already did.