{"version":"2.1.295","anchor":"auto-mode-classifier-context-is-tagged-with-a-new-marker-dr","canonical_anchor":"auto-mode-classifier-context-is-tagged-with-a-new-marker-dr","heading":"Auto mode safety check receives an extra marker","tier":"internal","area":"Auto Mode","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295\/e\/auto-mode-classifier-context-is-tagged-with-a-new-marker-dr","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295","markdown":"### Auto mode safety check receives an extra marker\n\nThe information Claude Code gives the auto mode safety check can now carry an extra marker, and a new prompt-screening step was added\n\n**Unclear.** It is not clear what the new marker or the prompt-screening step changes for someone using Claude Code.\n\n**What**\n\nIn auto mode, Claude Code uses a separate check, called a classifier, to decide whether an action can go ahead without asking you. The information passed to that check can now include an extra on-or-off marker in some cases. Claude Code also gained a new step for screening prompts in the interactive session.\n\n**Why**\n\nNeither change is described as altering what you see. They change what information is available when Claude Code decides how to handle a prompt or an action.\n\n- Flag `tengu_ax_screen_reader`: On for this account, and not off by default (read for one account on one subscription tier against v2.1.295; this account: on, anonymous baseline: on, compiled default: not a boolean we can read) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.\n- Area: Auto Mode\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}