You'll noticeTier: how much it should matter to you
3Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
Output LimitsArea: what it touches
ImprovementsKind: in v2.1.295,
ImprovementsSection of the release
What
Claude Code has a new message for when Claude reaches its output limit, the most text it may produce in one response, while still reasoning, before it has written any answer or used any tool. The message, beginning "Output token limit hit while you were still reasoning", tells Claude to reason much more briefly and carry on.
Why
It changes how Claude Code recovers when a response is cut off during the thinking stage, steering Claude toward finishing the task rather than stalling.