OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
MCP & Tooling

Computer Use Linux

Computer Use Linux is a Rust-based Linux desktop-control tool for AI agents, available as an MCP server, Pi native tools and a CLI. It exposes accessibility-tree inspection, screenshots, window targeting and input actions, with post-action verification.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Computer Use Linux — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an agent workflow needs structured inspection and control of a local Linux desktop, especially when accessibility-tree data and post-action checks are useful alongside screenshots and input.

Who it is for

  • Developers connecting an AI agent to a Linux desktop through MCP.
  • Pi Coding Agent users seeking native desktop-control tools.
  • Builders creating Linux desktop automation that uses accessibility data and input actions.

Documented uses

  • Inspect a desktop application through its accessibility tree.
  • Capture screenshots and identify or focus windows.
  • Send mouse and keyboard actions or invoke semantic accessibility actions.
  • Check desktop state after an agent interaction.
Before you choose

Requirements & limits

  • Installation routes include npm, Cargo, prebuilt Linux binaries, and integrations for several agent environments.
  • Desktop readiness depends on AT-SPI and an available input backend; optional setup can be required.
  • Input and window-targeting mechanisms vary across Wayland compositors and X11.
  • Prebuilt binaries are documented for Linux x86_64 and aarch64 and require glibc 2.28 or newer.

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

Install the CLI with npm and check desktop readiness

Install the command-line tool and use its readiness check to identify desktop setup needs.

Before you start

  • A Linux system
  • Node.js and npm
  • AT-SPI enabled and an input backend reported ready by the doctor command

Steps

  1. Install the package globally with the command below.

    Documented command
    npm install -g @agent-sh/computer-use-linux

  2. Run computer-use-linux doctor to check readiness.

  3. If needed, run computer-use-linux setup to enable the AT-SPI bridge.

  4. If needed for GNOME Wayland window targeting, run computer-use-linux setup-window-targeting.

Expected outcome

The CLI is installed and its doctor command reports desktop readiness; the setup commands address the listed accessibility or GNOME window-targeting needs.

Guide limitations

  • The npm installation does not replace required desktop system components.
  • Start ydotoold only when the doctor selects the ydotool fallback.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 November 2026
Original website title
agent-sh/computer-use-linux · Official homepage title · retained in its source language
How we review sourcesSuggest a correction