Every session already open on your Mac shows up as a card.
Tap in and you are in the same conversation, from the phone.
Talk instead of typing
Hold the mic and speak. The phone transcribes live so you can see the
words forming, then your Mac produces the final text and sends it. No
keyboard on a 6-inch screen.
Watch the tool calls land
Read, Edit, Bash, test output, diffs with line counts. The transcript
renders markdown and tables, so a plan or a comparison arrives readable
rather than as a wall of text.
Every session, as a grid
One card per Terminal window running Claude Code, each showing the
latest reply and whether it is working, waiting on you, or done. Swipe
between the grid, the queue, and your chats.
Watch localhost render as it is built
A real browser lives in the app, with tabs that keep their own history
and scroll position. Type 3000 or localhost:8787/admin
and it loads the dev server on your Mac. It goes through the same
authenticated route as everything else, including the tunnel, so
you can watch the agent's work appear from a train with no port
forwarding and no ngrok. The app lists which ports are listening, and
kills one that is stuck.
Mirror and drive the Mac
One tab over from the browser: the Mac's screen at three quality
tiers, with tap and keyboard input, for the dialog Claude cannot click
for you.
See what it is costing the machine
A vitals strip reads watts, temperature, CPU with a sparkline, and
memory used against total, straight off the Mac. Handy when a long run
starts cooking a laptop on a desk you are not sitting at.
Switch model and effort
Move a session to opus at xhigh in two taps.
The chip in the header shows which model answered and how much of the
recent output was thinking.
It tells you when a shell needs you
Claude stops and asks something, and your phone says
“jarvis needs you” — the folder name, so you know
which session, and a one-line summary of what it is actually asking
rather than a generic “needs input”. When a run finishes you get
“jarvis — tap to talk”.
Answer from the Lock Screen
The notification carries Claude's real choices as buttons. A yes-or-no
question gets Yes and No; a numbered menu gets its first
three options; a finished run gets Continue. Press and answer
without unlocking or opening anything.
Quiet about everything else
You only get pinged for sessions you started from the phone, so work
you are doing at the Mac never buzzes your pocket. It also stays silent
about the session already open on your screen. A morning digest
summarises what every project did overnight.
Live on the Lock Screen
A Live Activity ticks while the agent works, showing the session, its
elapsed time and token count in the Dynamic Island, with a home-screen
widget for the same. Your Mac pushes these itself.
Drop a file straight into the shell
Paperclip a screenshot, a photo, a log, a zip. It lands in that
session's working directory and the agent can read it immediately. Big
attachments get a progress bar and an hour to finish, so a video or an
archive over slow Wi-Fi does not time out at thirty seconds like it
used to.
Usage arcs wrapped around the Dynamic Island
Two concentric rings hug the island: the inner one is your 5-hour
limit with five hour-ticks etched in the track, the outer is your
7-day with seven day-ticks, each with a glowing head at the tip. Pace
evenly and the tip rides one tick per hour. Beside it, a countdown to
the next reset. Tap it for the pacing detail.
Queue work against the 5-hour reset
Hit your limit mid-thought and the work does not evaporate. Add
prompts to the queue and they fire on their own the moment the window
rolls over, into the session you picked. Come back to finished work
rather than to a blocked shell.
Pull the results back out
Browse everything the agent created or edited this session, per
session, and open or share any of it from the phone.
