{"version":"2.1.225","anchor":"auto-mode-classifier-failures-are-distinguished-from-classif","canonical_anchor":"auto-mode-classifier-failures-are-distinguished-from-classif","heading":"Auto-mode classifier failures are distinguished from classifier denials","tier":"notice","area":"Permissions","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.225\/e\/auto-mode-classifier-failures-are-distinguished-from-classif","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.225","markdown":"### Auto-mode classifier failures are distinguished from classifier denials\n\nWhen the auto-mode classifier can't decide, you no longer get a bogus denial message or permission-rule hint.\n\n**What**\n\nA permission decision that the classifier could not produce a verdict for is now tracked separately from a real denial, so telemetry and UI stop treating a classifier failure as a refusal.\n\n**Details**\n\n- The decision reason carries a new `noVerdict` flag.\n- Denial-reason bucketing keys off the flag instead of matching a message prefix; telemetry labels are assembled at runtime as `classifier:${e.classifier}:no-verdict` around the literal `:no-verdict`.\n- The UI denial kind `automode-parsing-error` is now chosen by `t.noVerdict`.\n- The 'add a Bash permission rule' hint and the post-denial hook message are both skipped for no-verdict cases.\n\n**Evidence**\n\n`noVerdict`\n\n- Area: Permissions\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 0\/5"}