{"version":"2.1.317","anchor":"auto-mode-classifier-servercleared-and-selfmending-flags-b","canonical_anchor":"auto-mode-classifier-servercleared-and-selfmending-flags-b","heading":"Auto mode's safety check records more detail about server verdicts","tier":"internal","area":"Auto Mode","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.317\/e\/auto-mode-classifier-servercleared-and-selfmending-flags-b","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.317","markdown":"### Auto mode's safety check records more detail about server verdicts\n\nAuto mode's command safety check now records more detail about server verdicts and about blocked actions\n\n**Unclear.** It is not clear whether these new markers change which actions auto mode allows or only what it records.\n\n**What**\n\nIn auto mode, Claude Code checks each action with a classifier, an automatic reviewer that decides whether the action can run without asking you. Some of these checks are made by Anthropic's servers. The record of each decision now carries more detail:\n\n- A decision can now note that the server cleared the action.\n\n- A check that gave no verdict because the conversation was too long is now marked as one that recovers on its own.\n\n- Claude Code now also checks whether the server did not flag an action it classified.\n\n- Blocked actions record more details for usage reporting.\n\n**Why**\n\nThese changes refine how auto mode handles verdicts from the server classifier.\n\n- Area: Auto Mode\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}