Why PhysiClaw Needs a Body

Why would a personal assistant need a robot at all? If it already lives on your phone, can’t it just talk to the apps directly?

It can’t. And the reason it can’t is the whole problem.

Every service you depend on — Amazon, Taobao, Uber Eats, Meituan — sits behind a wall. The data inside is their most valuable asset, and the one they guard most fiercely. They have no reason to open a door for your assistant, and every reason to keep it shut. An open API that lets your assistant roam freely across their service is the last thing they will ever ship.

So no clever software trick will save you. Any agent that tries to reach in through the front door hits the same wall as everyone else. For the foreseeable future, no one is going to hand you free, open access to the services that run your daily life.

If you can’t go through the wall, you go around it — through the one door they cannot lock.

That door is the screen. The same interface a human taps with a finger. They can shut their APIs, but they cannot shut the app to their own customers. The buttons, the menus, the screens you touch every day have to stay open — close those, and they have no business left.

So PhysiClaw uses that door. It works your phone the way you do. A camera for eyes. A stylus for a finger. The screen itself as its API. Whatever you can do by hand, it does by hand — through the exact same interface, in the exact same way.

This comes at a cost, and we won’t pretend otherwise. A task that takes software a few milliseconds takes PhysiClaw minutes — it has to look, move, tap, and look again. It is slower. Much slower.

But it buys something no API ever will: access that no one can revoke. PhysiClaw goes slow, and in return it goes anywhere. It trades speed for freedom.

That is the deal at the heart of PhysiClaw. The wall is built of software — permissions, tokens, terms of service. A hand is not. You cannot block a hand from touching a screen.

The data wall can be broken. Not with code — with hardware.