The closing response

Once every step in a turn is done, tCode gives a normal, plain-language response summarizing what it actually found and did, written the way a colleague would report back after actually finishing something, not a mechanical listing of steps taken.

Underneath that final message, a short activity line recaps the turn at a glance: which files were read, which were edited, and roughly how much changed, so scrolling back through the entire transcript is never the only way to see what happened.

This same output shape, tool headers, permission prompts inline, closing summary, is exactly what a remote-triggered turn produces too, described further in Pairing a machine.

The closing message is written fresh for that specific turn, not assembled from a template with blanks filled in; it reflects what genuinely happened, in whatever level of detail actually matters for that particular request, rather than a fixed-format report.

For a turn that made no changes at all, a purely investigative one, the closing message still follows the same pattern, summarizing what was found rather than what was done, so the shape of the output stays familiar even when nothing was written or edited.