Documentation
PhysiClaw gives an AI agent a physical body to operate a phone: a camera reads the screen and a stylus on a 3-axis arm taps it — no APIs, and nothing installed on the phone. These docs go from a parts list to a running agent, plus the tools and gestures it uses.
New here? Read the Introduction, then build one and run it.
Get started
Section titled “Get started” Introduction What PhysiClaw is and the problem it solves.
Installation Install the server and connect the hardware.
Quickstart Calibrate and run your first agent task.
Concepts
Section titled “Concepts” System architecture Agent → MCP server → cameras + arm → phone.
The control loop Perceive → decide → move → verify → touch.
Build & reference
Section titled “Build & reference” Bill of materials Every part for a ~$127 build.
MCP tools The six tools an agent calls.
Gestures How taps, long-presses, and swipes become G-code.