{"version":"2.1.314","anchor":"auto-mode-permission-retry-notice-reworded","canonical_anchor":"auto-mode-permission-retry-notice-reworded","heading":"Retrying a blocked action in auto mode now says it adds no permission rule","tier":"notice","area":"Auto Mode","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.314\/e\/auto-mode-permission-retry-notice-reworded","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.314","markdown":"### Retrying a blocked action in auto mode now says it adds no permission rule\n\nAfter you let Claude retry an action auto mode blocked, the notice now says Claude was told it may retry and that no rule was added\n\n**What**\n\nAuto mode is a permission mode where Claude Code decides on its own whether an action can go ahead. When you let Claude retry an action auto mode blocked, a notice appears in the conversation. It used to say \"Allowed\" followed by the action. It now says one of these:\n\n- \"Told Claude it may retry\", when Claude is told straight away.\n\n- \"At your next prompt, Claude will be told it may retry\", when the message has to wait until you next type something.\n\nBoth are followed by \"Auto mode checks it again; this adds no rule.\" Claude Code also records whether the message was held until your next prompt.\n\nIn remote sessions, the permission request shown when auto mode wants your decision now starts with \"Auto mode needs your confirmation for this command.\" (or \"tool\" for actions that are not shell commands), followed by the reason. Before, it started with \"An automated safety classifier requires confirmation:\".\n\n**Why**\n\nLetting Claude retry an action from the Recently denied list does not create a permission rule. Auto mode checks the action again, and the old \"Allowed\" wording could suggest otherwise.\n\n- Area: Auto Mode\n- Tier: You'll notice\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}