mcp-server

Control Linux with computer-use-linux

Name: Control Linux with computer-use-linux
Author: agent-sh

Gain desktop control over MCP with this tool, ideal for startup founders using Linux.

412 stars43 forksRustGuide quality 8/10Updated 8/13/2026100% free · open source

What it does

Computer-use-linux allows startup founders to gain desktop control over MCP on Linux systems, leveraging AT-SPI, GNOME Shell, Wayland portals, and ydotool.

When to use it

•When automating GUI interactions on Linux desktops
•Integrating Linux desktop functionality with MCP server
•Testing or simulating user interactions on Linux systems

Ready-to-paste prompt

cargo run --example ydotool -- -c 'move 100 100' to move the mouse cursor to coordinates (100, 100)

Heads up: Ensure you have the necessary dependencies installed, including cargo and the Rust compiler, and that your system supports AT-SPI, GNOME Shell, and Wayland portals

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/computer-use-linux && curl -fsSL https://workflowstacks.com/api/skills/computer-use-linux/claude-skill -o ~/.claude/skills/computer-use-linux/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How Control Linux with computer-use-linux works

Codeflow

Free to inspect

Control Linux with computer-use-linux is a large Rust project (~17k lines across 43 code files, plus 8 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, MIT license, has a test suite.

Size

Large codebase

~17k lines · 43 code files · ~2 h to skim

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Rust · Node.js

No API keys detected

Rust 57%JavaScript 32%TypeScript 6%Shell 3%Python 2%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
src/cli.rs
Where the program starts running
3
Cargo.toml
Dependencies and the commands it exposes
4
examples/atspi_probe.rs
A worked example — copy this to get going

What's in each folder

src/Core code — the actual logic28 files

skills/Prompts, skills & agent definitions3 files

examples/Examples you can copy1 files

pi/Folder15 files

scripts/Helper scripts6 files

npm/Folder3 files

gnome-shell-extension/Folder2 files

tests/Tests — proof it works3 files

READMEHas testsDocumentedCI checksExamples includedMIT licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/computer-use-linux && curl -fsSL https://workflowstacks.com/api/skills/computer-use-linux/claude-skill -o ~/.claude/skills/computer-use-linux/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

agent-sh

Language

Rust