{"version":"2.1.239","anchor":"bash-commands-from-a-remote-agent-always-ask-a-person","canonical_anchor":"bash-commands-from-a-remote-agent-always-ask-a-person","heading":"Bash commands from a remote agent always ask a person","tier":"notice","area":"Permissions","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.239\/e\/bash-commands-from-a-remote-agent-always-ask-a-person","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.239","markdown":"### Bash commands from a remote agent always ask a person\n\nBash commands from a remote agent now always wait for a human and never run sandboxed.\n\n**What**\n\nCommands that arrive from a remote agent are no longer eligible for sandboxed execution and no longer auto-approvable by the permission classifier, so they run unsandboxed and wait for a human answer.\n\n**Details**\n\n- Both the Bash and PowerShell permission hooks require the request source to not be a remote agent before treating a command as sandboxable.\n- The auto-approval path treats a remote-agent source as not classifier-approvable, the same way it treats a request whose ask was capped by organization policy.\n- No flag or setting gates this.\n\n**Evidence**\n\n`e.requestSource?.type !== \"remote-agent\"`\n\n- Area: Permissions\n- Tier: You'll notice\n- Useful: 4\/5\n- Signal: 4\/5"}