Claude Code notifications on your Apple Watch

Written by the maker of Tapproval, one of the options below, so read it with that in mind. Every claim about another product links to its source; facts as of 25 October 2026.

Claude Code stops whenever it needs you: a command to allow, a question to answer, a plan to approve. If you have stepped away, the session simply waits. There are three ways to have your Apple Watch tell you. They differ in one thing that matters more than it sounds: whether the watch can only tell you, or can also answer.

1. Remote Control's push, through your iPhone

Claude Code's own Remote Control can push a notification to the Claude app on your phone, "typically when a long-running task finishes or when it needs a decision from you". To turn it on:

• Install the Claude app on your iPhone and sign in with the account you use for Claude Code.
• Start Remote Control in the session: /remote-control.
• Type /config and turn on Push when actions required (permission prompts and questions), Push when Claude decides, or both.

Remote Control needs a Pro, Max, Team or Enterprise plan; an API key does not work. Your watch shows the Claude app's notification the way it shows any iPhone app's: while the iPhone is locked or asleep. The Claude app runs on iPhone and iPad and has no Apple Watch app, so the prompt itself is answered on the phone.

Good fit if you already pay for Claude and carry your phone. It is built in and installs nothing on the watch.

2. A notification hook that pings your phone

Claude Code runs a Notification hook when it is waiting for you. A short script there can send a push through a service such as ntfy or Pushover, which your watch then shows like any other phone notification. Small open-source projects do exactly this, for example scottmatthews7/claude-watch, which notifies when a task finishes.

It works with any plan and any computer. Most such scripts only tell you, and you still go to the terminal or the phone. Some add Allow and Deny buttons to the notification, such as claude-remote-approver.

3. A watch app that answers

The third kind puts the prompt itself on the watch and sends your answer back. On the computer, a PermissionRequest hook hands each prompt to a small relay; the watch app shows it, and one tap answers. Tapproval works this way: each prompt arrives marked with one of five risk tiers, worked out on your computer, so a force-push to main shows as CRITICAL before you tap. The first 100 answers are free, then $5 once.

Other watch projects exist, several of them source code you build yourself. The comparison page sets them side by side, with price, account and server for each.

When your iPhone is not with you

This is where the three part ways. Apple routes an iPhone app's notifications to the watch only while the two are connected: "when your devices are disconnected, your notifications will appear on your iPhone instead of your Apple Watch." Leave the phone at home and options 1 and 2 go quiet on the wrist.

A watch app can reach your computer on its own over the internet, on Wi-Fi or cellular. Tapproval does that through a private travel address and through your own iCloud, so prompts still arrive and answers still land with the phone switched off. The setup guide covers how it finds the computer.

Using Remote Control and a watch app together

They combine. The one thing to know: with Push when actions required on, the Claude app also pushes for prompts the watch app has already answered, so one prompt can buzz twice (claude-code#96126). The setup guide explains which switch keeps one buzz, and what that costs.

Side by side

Using Codex too? Codex notifications on Apple Watch covers its four ways.

Get Tapproval on the App Store Setup guide