We don’t want to stay in the IDE right now—we just want to understand what the Agent is doing and why it’s doing it
All these Agent-based programming TUIs
are only a phase.
A few months from now,
most developers will go back
to the embrace of GUI and IDE.
But right now,
I don’t want to edit code in the IDE at all.
I don’t want to stare at every line of code.
I just want to make sure
I understand what the agent is doing,
and why it has to do it this way.
A desktop version of CodeX with multi-agent parallelism might be the best interaction at this stage
