Skip to content
PhysiClaw PhysiClaw Doc
English

Introduction

PhysiClaw is a small desktop robot that gives an AI agent a physical body to operate a phone. It watches the screen through a camera and touches the glass with a capacitive stylus on a 3-axis arm. There is no app to install on the phone, no API to integrate, and no account to connect — to the phone, PhysiClaw is indistinguishable from a human finger.

The problem

Section titled “The problem”

Software agents that “use your phone” usually depend on one of three things, and each is a wall:

APIs & OAuth

Every service needs its own integration, keys, and consent screens. New app, new wall.

Accessibility hooks

Screen readers and automation frameworks are blockable, detectable, and break on redesigns.

Jailbreaks

Rooting a device to inject taps is fragile, unsafe, and off-limits for most people.

The PhysiClaw way

Section titled “The PhysiClaw way”

Instead of a software bridge, give the agent presence. The loop is deliberately simple:

  1. A top camera photographs the screen.
  2. The agent decides the next action from the image.
  3. The arm moves the stylus and a side camera confirms the tip is on target.
  4. The stylus touches, then retracts — and the loop repeats.

Because the only thing touching the phone is a stylus tip, any app works — iOS or Android, your bank, a delivery app, a chat thread — with zero per-app setup.

What’s next

Section titled “What’s next”