{"version":"2.1.288","anchor":"hook-output-with-permissiondecision-ask-detected","canonical_anchor":"hook-output-with-permissiondecision-ask-detected","heading":"Claude Code picks out hook replies that ask for your permission","tier":"internal","area":"Hooks","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.288\/e\/hook-output-with-permissiondecision-ask-detected","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.288","markdown":"### Claude Code picks out hook replies that ask for your permission\n\nClaude Code now singles out hook output whose permission decision is \"ask\", unless the hook also said to stop\n\n**Unclear.** What Claude Code does with replies that pass this check is not known.\n\n**What**\n\nA hook is your own command that Claude Code runs at set moments, and it can reply with a permission decision about a tool call. Claude Code now has a dedicated check that spots a reply whose `permissionDecision` is \"ask\", meaning you should be prompted. A reply that also says not to continue is left out of that check.\n\n**Why**\n\nThis suggests hook replies of \"ask\" may get handling of their own.\n\n- Area: Hooks\n- Names: `permissionDecision`\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}