{"version":"2.1.296","anchor":"new-recovery-message-for-an-empty-response-that-hit-the-outp","canonical_anchor":"new-recovery-message-for-an-empty-response-that-hit-the-outp","heading":"Recovery message for replies that hit the output limit without showing anything","tier":"internal","area":"Output Limits","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.296\/e\/new-recovery-message-for-an-empty-response-that-hit-the-outp","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.296","markdown":"### Recovery message for replies that hit the output limit without showing anything\n\nA new built-in message asks the model to continue briefly when a reply reached the output limit with no visible output\n\n**What**\n\nThe output limit is the most text the model can produce in a single reply. Claude Code has a new built-in message for the case where a reply hit that limit without showing you anything. It reads: \"Your previous response reached the output limit with no visible output. Please continue and produce a brief user-visible response or tool call right away.\" A tool call is the model asking Claude Code to run one of its tools, such as reading a file.\n\n**Why**\n\nIt gives Claude Code a way to recover from a turn that used up its whole output allowance before producing anything visible.\n\n- Area: Output Limits\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}