# Hanzo Computer — Hanzo AI

> An agent that takes a screenshot, decides where to click, and clicks. No API and no integration for the apps it drives — it uses the screen, like you do.

computer.hanzo.ai

# Hanzo Computer

An agent that uses the computer in front of it

It takes a screenshot, decides where to click, and clicks — the same inputs and outputs a person has. Say what you want in a sentence and it works through the apps already on the machine, with no API and no integration for any of them. Install it with pip, run operate, and type an objective.

[Try Hanzo Computer](https://computer.hanzo.ai)[Documentation](https://docs.hanzo.ai/docs/skills/hanzo-computer)

## If a person can click it, so can this

The screen is the interface. That is the whole trick, and it is why the app you need to automate does not have to know anything about it.

### It reads pixels

No DOM, no accessibility tree, no selectors to maintain. A native app, a canvas, a remote desktop and a web page all look the same to it — like a screenshot.

### Mouse and keyboard

Mac, Windows, and Linux with an X server. On macOS the terminal asks for Screen Recording and Accessibility the first time, because those are exactly the permissions this needs.

### OCR when clicking is hard

Add -with-ocr to the model and it reads text off the screen to build a map of clickable elements — the fix for a button a model can see but keeps missing by twenty pixels.

### Set-of-Mark prompting

Add -with-som and the screenshot arrives pre-labelled, so the model names a marked element instead of guessing a coordinate.

### Say it out loud

operate --voice takes the objective by microphone. Needs the audio extras and portaudio installed; it is not in the base install.

### It has whatever you have

This runs as you, on your machine, with your logged-in sessions. Give it a fresh user account or a VM before you give it anything that matters — and remember that text on a web page can try to redirect it.

## For the software with no API

The vendor portal that exports one CSV at a time. The desktop tool from 2009. The internal app nobody will reopen. Those.

[Try Hanzo Computer](https://computer.hanzo.ai)[View on GitHub](https://github.com/hanzoai/computer)

## Get started with Hanzo Computer

[Read the docs](https://docs.hanzo.ai/docs/skills/hanzo-computer)[View on GitHub](https://github.com/hanzoai/computer)

## Open source

License: Apache-2.0[hanzoai/computer](https://github.com/hanzoai/computer)

## Get Computer

Computer-use for AI

[Deploy to Cloud](https://console.hanzo.ai/deploy)[Self-host](https://docs.hanzo.ai/docs/skills/hanzo-computer)