Introduction
Why NZT
A coding agent can get to the wrong place very fast. See what changes with NZT: analysis before code, reviewable units and honest reports of what happened.
An agent can get to the wrong place very fast. NZT changes how it works, not what it knows.
Without NZT and with NZT
| Without NZT | With NZT |
|---|---|
| It writes code before anyone has defined what it should do. | Functional analysis comes first, and anything undefined becomes a question, not an assumption. |
| It starts and does not stop until something breaks. | It cuts the work into reviewable units and stops to wait for you at the end of each one. |
| It runs out of context and you lose the thread. | The plan and the state live in Plan/state.json: you resume without explaining it all again. |
| It picks the pattern it likes today, different from yesterday’s. | The stack document decides, and an installed folder is not an authorization. |
| It tells you it works. | It reports what happened: the test that failed, the step it skipped, the assumption it made. |
What you gain
What you gain is not better code. You can review what it did, pick up where you left off, and avoid finding out at the end that it understood something else.