I give Claude Code longer and longer tasks. Agents got smarter, so I can set a task, switch to something else, and come back later.

I set up a simple alarm with Claude Code hooks, shell commands that fire on lifecycle events, all calling the macOS say command:

Here’s my ~/.claude/settings.json:

{
  "hooks": {
    "Stop": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "say -v Zarvox \"Human, I'm done\"",
            "async": true
          }
        ]
      }
    ],
    "Notification": [
      {
        "matcher": "permission_prompt",
        "hooks": [
          {
            "type": "command",
            "command": "say -v Zarvox 'Human, I need your permission'",
            "async": true
          }
        ]
      },
      {
        "matcher": "elicitation_dialog",
        "hooks": [
          {
            "type": "command",
            "command": "say -v Zarvox 'Human, I need more information'",
            "async": true
          }
        ]
      }
    ]
  }
}

I use two notification matchers: permission_prompt and elicitation_dialog. I used to have an idle_prompt matcher, but I removed it quite fast. When you run multiple agents, after a while they all start complaining they’re bored.

Zarvox is the robotic voice. Run say -v '?' | grep en_US to see what’s available. Whisper specifically is a nice one, but scary as hell if you work late at night.

The "async": true flag tells Claude Code to fire the hook and keep working without waiting for say to finish. For a quick say command it barely matters, but without it Claude technically pauses until the command completes.

If you want something more sophisticated, you may try Peon Ping, a tool that solves the same problem with Warcraft III voice lines, desktop notifications, and a desktop orc tamagotchi. I tried it, appreciated the concept, but still, uninstalled it pretty quickly and replaced it with that say command.