New agentic code review workflow unlocked! Before: You review a diff. Drop some comments. The agent makes changes but they're mixed with old changes. So you re-read everything. Give more feedback. Repeat. After: Diffs only show what's new since your last review, in realtime.

Learn more in - I'm looking forward to your feedback!

This looks amazing, way better than the current flow I am using. I'll give tidewave a try this week. Are you considering adding a tab like that for tests too? Or things that you want to run after every AI change (such as formatting, running credo, etc)

Those workflows (formatting, running credo, etc) can be added/handled directly in your coding agent!

This is not a problem if you are using jj. I run "jj new" before prompting agent to do additional work and it only contains the new diff. I squash it back to the original revision if I am happy with it. Simple and powerful.

Partially. You could also do the same if you always git commit before asking the agent but that only works wholesale. If you have done only a partial review, now you need to git add files manually or call jj split. It is even more annoying if you have reviewed only certain hunks in a large file.

The realtime delta is the key piece. One extra layer that would help review loops even more: map each new hunk back to the original review comment or intent so you can see whether the agent actually resolved the feedback or just changed nearby code.