OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Browser Agents

Open Computer Use

Open Computer Use is a Python application that runs an AI-controlled Linux desktop in an E2B Desktop Sandbox. The agent uses keyboard, mouse and shell actions, streams the desktop display to a client computer, and lets a user pause and prompt it.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Open Computer Use — OpenViST project profile in Browser Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an AI task needs desktop-level controls and a visible, interruptible session rather than browser-only interaction.

Who it is for

  • Developers exploring LLM-controlled desktop automation.
  • Teams or researchers who want to observe and intervene in an agent's computer-use session.
  • Developers who want to configure a computer-use agent with supported vision, grounding or action models.

Documented uses

  • Give an AI agent a desktop environment in which it can act through keyboard, mouse and shell controls.
  • Watch the sandbox display while the agent works and pause it to provide another prompt.
  • Choose and configure supported models for grounding, vision and action.
Before you choose

Requirements & limits

The documented setup requires an E2B API key and credentials for the selected LLM provider, with Hugging Face Spaces identified as not requiring an API key.

  • The application depends on E2B Desktop Sandbox for its cloud desktop environment.
  • The README lists more than ten LLM options, but model capabilities differ across grounding, vision and action roles.
  • The live display stream and pause-and-prompt control support human intervention during use.
  • The project uses Ubuntu and says it is designed for other operating systems, but does not identify a tested alternative.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the Open Computer Use interface

Install the documented prerequisites, configure access keys and start the web interface.

Before you start

  • Python 3.10 or later
  • Git
  • Poetry and FFmpeg
  • An E2B API key
  • An API key for the selected LLM provider; Hugging Face Spaces do not require an API key

Steps

  1. Install Poetry and FFmpeg using the README's brew install poetry ffmpeg command.

  2. Clone the e2b-dev/open-computer-use repository and enter its project directory.

  3. Create a .env file containing E2B_API_KEY and the credentials for the provider or providers selected in config.py.

  4. Run poetry install, then poetry run start to start the agent interface.

Expected outcome

The README says the agent opens and prompts for an instruction, with the sandbox display stream expected to appear a few seconds after the Python program starts.

Guide limitations

  • The documented installation command uses Homebrew.
  • A Hugging Face token is specified to bypass Gradio rate limits when using Hugging Face Spaces.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
e2b-dev/open-computer-use · Official homepage title · retained in its source language
How we review sourcesSuggest a correction