Runs on your Mac

Nimrobo has a macOS app. Point it at one number you own. The agent tries a change, checks whether the number moved, and keeps what worked.

macOS 12+ · Apple silicon and Intel · Sign in with ChatGPT Plus or Pro, or bring your own API key

Dashboard · Outcome
onboarding activation
Add skillChatRun next action
Outcome
Levers
Hypotheses
Experiments
Sessions
Files
Settings
North star
67.1%
0.3 vs 20/07/2026, 06:00
activation_rate_7d · increase is better · read 27/07/2026, 06:00
activation_rate_7d67.1%
11/06/202627/07/2026
▸ View exact readings
support_tickets_7d41 tickets
11/06/202627/07/2026
time_to_first_action_p504.2 min
11/06/202627/07/2026
4 open bets · 3 open experiments · 9 shipped artifacts
Outcome pulse
3experiments open
4hypotheses open
9artifacts shipped
Last outcome reading 27/07/2026, 06:00. Next collection 03/08/2026, 06:00.
Last 5 sessions
outcome actionShip checklist variant B to the 10% cohort27/07/2026, 09:12runningOpen
updateFold rep-3 reward into the sample-workspace verdict26/07/2026, 18:40completedOpen
outcome actionDraft the day-1 stalled-team email25/07/2026, 11:02waitingOpen
outcome actionRead activation for week 29 and log the snapshot24/07/2026, 06:04completedOpen
chatWhy did tooltips lose to the checklist?23/07/2026, 20:31completedOpen
01WHAT'S MISSING

Why can't I just ask ChatGPT?

You can, and it will give you ten good ideas. It will even remember the conversation. What it never learns is the result. Nobody tells it which ideas you shipped, or whether the number moved. So next month you get the same ten ideas, with the same confidence.

So you accumulate output. Never knowledge.

Nimrobo solves this.

Your number · what caused each move?
02The harness

The agent does the work. The harness keeps score.

Point Nimrobo at one number. Every run, it reads back what its earlier runs proved and disproved about that number, decides what's worth trying next, and states what it expects before it acts. The real result settles that prediction and joins the record the next run reads from.

Dashboard · Outcome
onboarding activation
Add skillChatRun next action
Outcome
Levers
Hypotheses
Experiments
Sessions
Files
Settings
North star
67.1%
0.3 vs 20/07/2026, 06:00
activation_rate_7d · increase is better · read 27/07/2026, 06:00
activation_rate_7d67.1%
11/06/202627/07/2026
▸ View exact readings
support_tickets_7d41 tickets
11/06/202627/07/2026
time_to_first_action_p504.2 min
11/06/202627/07/2026
4 open bets · 3 open experiments · 9 shipped artifacts
Outcome pulse
3experiments open
4hypotheses open
9artifacts shipped
Last outcome reading 27/07/2026, 06:00. Next collection 03/08/2026, 06:00.
Last 5 sessions
outcome actionShip checklist variant B to the 10% cohort27/07/2026, 09:12runningOpen
updateFold rep-3 reward into the sample-workspace verdict26/07/2026, 18:40completedOpen
outcome actionDraft the day-1 stalled-team email25/07/2026, 11:02waitingOpen
outcome actionRead activation for week 29 and log the snapshot24/07/2026, 06:04completedOpen
chatWhy did tooltips lose to the checklist?23/07/2026, 20:31completedOpen
  • north star + trend
  • guardrails
  • open experiments
03Control

An agent on your Mac, on a leash you hold.

The agent runs shell commands on your Mac. Three separate settings decide how far it can go on its own: which files and networks a command can reach, how often it stops to ask you, and how fast you can kill it. You set all three.

The blast radius

What can any one command actually reach?

By default, almost nothing outside the folder you point it at. Reaching further is a ladder: every rung up costs an approval.

standardSilent

Outcome folder + /tmp · network blocked · secrets sealed

elevatedAsks every call

Writes anywhere · network open · secrets still sealed

no-sandboxAPPROVAL EVERY TIMEAsks every call

Full machine access · your approval on every call

Not a promise the model makes: a macOS Seatbelt profile the kernel enforces.

Sandboxed commands run inside Seatbelt, Apple's own OS sandbox. A blocked write fails at the operating system, not because the agent chose to behave. A project config can widen a run's room but can't un-protect your secrets, and sudo, rm -rf /, and sandbox escapes are blocked outright.

The dial

How closely do I have to watch it?

A separate setting from the sandbox: one you move, from fewest-approvals to approve-everything to read-only.

DefaultThe fewest approvals. Elevated and no-sandbox calls still ask.
ManualEvery edit, write, and command asks. Nothing remembered.
PlanRead-only. It proposes a plan; edits stay blocked until you approve.

Switch mid-run and it reaches subagents already running, not just the next one.

Stepping in

Can I get in its way while it's running?

At any second. It's never a black box you have to wait out.

Steer

Type while it's running. It picks you up at the end of the current step, not after the whole turn finishes.

It asks you

When a call is yours, it stops and asks. Dismiss it and the turn halts.

Stop

One button aborts the run and kills the process group.

04Templates

Start with the number you own.

An outcome loop needs a number you can move and can measure. Four templates ship ready to run, each one naming the metric, the system it is read from, and the line it will not cross to move it.

07FAQ

The fine print, without the fine print.

Start your first outcome.

Pick one number. Nimrobo works on it every day and builds the record of what actually moves it.

macOS · Apple silicon and Intel · No credit card