{"version":"2.1.300","anchor":"agent-spawn-hook-rewrite-and-remote-launch-fixes","canonical_anchor":"agent-spawn-hook-rewrite-and-remote-launch-fixes","heading":"Two fixes when launching agents in the cloud","tier":"notice","area":"Cloud Sessions","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300\/e\/agent-spawn-hook-rewrite-and-remote-launch-fixes","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300","markdown":"### Two fixes when launching agents in the cloud\n\nLaunching a cloud agent no longer starts sign-in recovery for errors that already explain themselves, and remote launches return the right result\n\n**What**\n\nTwo small fixes apply when Claude Code launches an agent in the cloud:\n\n- Sign-in recovery now starts only for a \"not logged in\" failure that has no message of its own. A failure that carries its own message no longer sends you into sign-in.\n\n- A successful remote launch now returns the correct result data.\n\n**Why**\n\nIn practice this means fewer unneeded sign-in prompts when a cloud launch fails for some other reason.\n\n- Area: Cloud Sessions\n- Tier: You'll notice\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}