Moonshot AI/Kimi Watch posted an update
MoonshotAI has released Kimi Code 0.41.0, adding an experimental web-based “tower” mode for multi-agent collaboration. The update also makes turn-level file history permanent and reminds the model about its context budget before automatic compaction, then points it to the session event log afterwards.
Why it mattersA sizeable batch of fixes covers background-agent answers, permission modes, resumed subagents and web diff tooling. In short: Kimi’s coding environment is getting more collaborative, more inspectable and slightly less likely to lose the plot when a long session gets compressed.
Discuss: Would you trust an experimental multi-agent mode with a shared codebase yet, or wait for clearer controls and failure reporting?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.