{"version":"2.1.307","anchor":"permission-explanation-shows-sandbox-override-reason-and-a","canonical_anchor":"permission-explanation-shows-sandbox-override-reason-and-a","heading":"Permission explanations now name a host's ask and sandbox overrides","tier":"notice","area":"Permissions","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.307\/e\/permission-explanation-shows-sandbox-override-reason-and-a","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.307","markdown":"### Permission explanations now name a host's ask and sandbox overrides\n\nPermission explanations now tell an ask from a host apart from an ask cap, and expand when a sandbox override caused the decision\n\n**What**\n\nWhen Claude Code explains why it is asking your permission, it labels where the ask came from. That label now tells two sources apart:\n\n- \"ask cap\", a limit that forces an ask\n\n- \"host's ask\", an ask from the program hosting Claude Code\n\nBefore, every case fell back to \"ask cap\". The explanation now also expands when a sandbox override caused the decision; a sandbox is the restricted space commands run in. An approver field has been added to the list of permission details.\n\n**Why**\n\nPermission prompts say more clearly where they came from, which helps most when Claude Code runs inside another program.\n\n- Area: Permissions\n- Tier: You'll notice\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}