Deskwright

Deskwright

Computer use for AI agents on GNOME Wayland.

Let your coding agent use your Linux desktop. Or give it one of its own, so it stops stealing your mouse.

Install • What it can do • A desktop of its own • When something's off • Security

Claude Code can edit your files and run your commands. It can't open GIMP, click a button in a settings dialog, or read what a native app is showing you. On macOS and Windows it can, through computer use. On Linux that's still on Anthropic's list.

This fills the gap on GNOME. It's an MCP server plus a small shell extension, and it hands an agent the actual desktop: launching apps, reading widgets, clicking, typing, dragging, window management, OCR and screen recording. Any MCP client can drive it. Claude Code, Codex, Cursor, your own script.

The part people tend to like most: DESKWRIGHT_SESSION=headless runs all of it on a virtual monitor that isn't on any of your screens. Your agent gets a real GNOME desktop to work on, and it never takes your focus.

Linux hybrid beta: 0.2.0b1

The beta adds guarded native controls, exact multi-control queries, adaptive visual fallback and shorter verified execution batches. It supports GNOME Wayland; Windows is not implemented. See the interaction guide and measurements and limits.

For a new installation:

pipx install --system-site-packages 'deskwright==0.2.0b1'

deskwright-setup

For an existing pipx installation, pipx upgrade --pip-args='--pre' deskwright keeps its existing system-site-packages setting. Run deskwright-setup to copy the updated extension, then log out and back in if setup reports a changed extension. Restart the MCP connection to load the updated Python tools. A connection restart alone cannot reload GNOME Shell's extension.

Why this exists

Wayland deliberately stops an application from seeing or touching any other window. That's a good rule, and it's why xdotool, wmctrl and grim either do nothing on GNOME or fail in ways that look like your own mistake. Every Linux computer-use project I found either shipped a Docker container running X11, or quietly assumed X11 and broke.

So I went looking for what GNOME actually permits, and it turns out to be quite a lot, just not where anyone looks. Mutter answers D-Bus. AT-SPI, the accessibility layer built for screen readers, exposes every real widget in every running app with its name, its role and the action it performs. And a shell extension runs inside gnome-shell itself, where the rest lives.

That middle one matters more than it sounds. Pressing a widget's own accessibility action isn't a nicer way to click. It's a different thing: it can't miss, it survives the window moving, and it needs no pointer at all. An agent working this way stops guessing at coordinates, and stops taking a screenshot after every action to find out what happened.

Install

You need GNOME Shell on Wayland. Check with echo $XDG_CURRENT_DESKTOP $XDG_SESSION_TYPE, which should mention GNOME and wayland.

The lazy way: hand it to your agent

Open Claude Code, or Codex, or whatever you use, and say:

Set up https://github.com/tristanmuzzu/deskwright on this machine,

follow the AGENTS.md.

AGENTS.md is a runbook written for agents. Every command, how to check each one worked, the right package names for Debian, Fedora and Arch, and the handful of things that go wrong with their fixes. Your agent will ask you for a sudo password once and tell you to log out once. That's your whole involvement.

By hand: two commands and a logout

pipx install --system-site-packages deskwright

deskwright-setup

deskwright-setup narrates every step. It turns on the accessibility flag, installs the bundled shell extension, and tells you about any missing system package with the right install line for your distro, so you're never guessing at package names. It never runs sudo itself.

Then log out and log back in. Once. There's no way around this one: on Wayland, gnome-shell only picks up an extension at session start.

Last step, point your client at it:

claude mcp add deskwright --scope user -- deskwright

That's it. Ask your agent to open an app and it will.

Why --system-site-packages , and what breaks without it

PyGObject publishes no wheels to PyPI. It's a distro package everywhere (python3-gi, python3-gobject, python-gobject), so an isolated venv can't import it. Leave the flag off and you get a server that starts cleanly and then dies on the first click with input_backend_failed.

uvx has no equivalent flag, which is the only reason this says pipx. If you'd rather skip pipx, a plain python3 -m venv --system-site-packages followed by pip install deskwright works identically.

As a Claude Code plugin instead

The repo is also a plugin marketplace. It registers the server plus a skill that teaches an agent which tool to reach for first.

claude plugin marketplace add tristanmuzzu/deskwright

claude plugin install deskwright@deskwright

The plugin runs the server from its own checkout, so there's no pip install. You still need the system packages and the extension, so run the setup out of the checkout Claude Code cloned for you, once:

~/.claude/plugins/marketplaces/deskwright/bin/deskwright-setup

Then log out and back in, same as above.

From a clone, if you want to hack on it

git clone https://github.com/tristanmuzzu/deskwright

cd deskwright bin/deskwright-setup ./mcp_server.py --self-test

./mcp_server.py is the same entry point as the deskwright command, by the path older registrations already point at. CONTRIBUTING.md has the layout and which test suites need a real session.

Check it works

DESKWRIGHT_SESSION=headless deskwright --self-test

You want 18/18 passed. It runs on a virtual monitor rather than your screen, so it's safe to run while you're working. The first run takes about 20 seconds because it has to start a second gnome-shell.

Drop DESKWRIGHT_SESSION=headless and it tests your real desktop instead. Do that one while you're looking at the screen: the self-test injects real input, because it's checking the guards that refuse dangerous key combinations.

When something's wrong, ask your agent to call desktop_health. It answers in one line whether this desktop is usable, then says which mechanisms work right now and what each of them will actually do.

What it can do

41 tools. Each returns enough that you do not need a second call to work out what happened. The order below is the order an agent should reach for them. Use native controls and exact queries where useful, with screenshots and ordinary input for visual work or incomplete accessibility.

A second desktop it uses while you work

An agent that needs your screen is only half useful. deskwright-headless starts a separate GNOME session on a virtual monitor, with its own session bus, its own gnome-shell --headless and its own runtime directory. A server pinned to it drives that desktop with the same 41 tools while you keep the physical one.

deskwright-headless start                    # about 200 MB of gnome-shell, idempotent

deskwright-headless status # live