{"version":"2.1.287","anchor":"auto-mode-classifier-new-messages-for-hook-rewritten-input","canonical_anchor":"auto-mode-classifier-new-messages-for-hook-rewritten-input","heading":"Clearer auto mode messages when a hook rewrites a tool call","tier":"notice","area":"Auto Mode","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.287\/e\/auto-mode-classifier-new-messages-for-hook-rewritten-input","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.287","markdown":"### Clearer auto mode messages when a hook rewrites a tool call\n\nAuto mode now tells Claude when a `PreToolUse` hook rewrote a call, so it can retry once and then explain the cause to you\n\n**Unclear.** It is not clear what condition triggers the repeat-once message.\n\n**What**\n\nIn auto mode, a classifier, an automatic check, decides whether Claude's tool calls can run without asking you. When it refuses a call, Claude Code tells Claude why. Two cases now have their own messages:\n\n- when a hook changed the call's input after Claude wrote it, Claude is told to retry once and, if refused again, to tell you that a `PreToolUse` hook or mod is rewriting the call\n\n- when the call was not one Claude requested, and an extra condition is met, Claude is told to repeat the identical call once\n\nPreviously the second case got a generic message and there was no rewrite case.\n\n**Why**\n\nIf a hook of yours rewrites tool calls, auto mode refusals caused by it are now explained to you rather than left puzzling.\n\n- Area: Auto Mode\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 2\/5"}