{"version":"2.1.248","anchor":"resuming-a-stranded-agent-uses-the-current-permission-callba","canonical_anchor":"resuming-a-stranded-agent-uses-the-current-permission-callba","heading":"Resuming a stranded agent uses the current permission callback","tier":"notice","area":"Permissions","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248\/e\/resuming-a-stranded-agent-uses-the-current-permission-callba","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248","markdown":"### Resuming a stranded agent uses the current permission callback\n\nResuming a stalled agent uses your current permission state rather than an old snapshot.\n\n**What**\n\nWhen an agent with queued messages is resumed, the permission-check callback is now fetched at resume time rather than being the one captured when the UI first rendered, so it reflects current permission state.\n\n**Evidence**\n\n`Failed to deliver queued message to agent: `\n\n- Area: Permissions\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 2\/5"}