agent-desktop
Computer use CLI for AI agents
Runssource: GitHubRustApache-2.0commit d3dc190afbf3
Rust, Apache-2.0 licensed. The project labels itself: accessibility, accessibility api, ai agents, automation, cli, computer use, computer use agent and computer use agents.
agent-desktop runs. An Argusic agent installed it in 9 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
The npm package and source release binary run as agent-desktop 1.1.0; all 2,400 executed Rust tests passed, and the real GTK fixture was observed, clicked, typed into, and read back successfully through the Linux accessibility tree.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page. Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- Linux source build initially lacked AT-SPI/GTK session support, so accessibility commands returned PLATFORM_NOT_SUPPORTED5 minutes
- Rust toolchain was absent from PATH2 minutes
- npm install -g agent-desktop failed with EACCES because npm prefix /usr/local is owned by root1 minute
- Install time
- 9 minutes
- Cold machine to finish
- 13 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- cargo build passed; cargo fmt --check, cargo clippy -D warnings, architecture/source checks exited 0; 2,169 library tests passed (233 ignored) and 231 binary tests passed; live GTK fixture snapshot returned ok:true complete:true ref_count:37, click exited 0 and logged zoo-button activate, type returned delivered_verified with hello-argusic, and the next snapshot read that value back
- Model tokens used
- 243,162
- Exact commit tested
- d3dc190afbf3
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
- Real GTK end-to-end: `list-apps` found the app, snapshot returned `ok:true` with 37 refs, `click`… - Git state is clean; no repository changes were made. tokens used 243,162 ## Result Standard `npm install -g agent-desktop` failed because `/usr/local` is root-owned. I fixed it with … - Installed npm package version `1.1.0` responded with the expected `version` envelope. - Source release binary built successfully: 7,103,032 bytes. - `cargo fmt` and `cargo clippy -D warnings` exited 0; architecture and source checks passed. - Test suites passed: 2,169 library tests (233 ignored) plus 231 binary tests. - Real GTK end-to-end: `list-apps` found the app, snapshot returned `ok:true` with 37 refs, `click`… - Git state is clean; no repository changes were made.
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 9 minutes, faster than the median of the 10 comparable projects Argusic has measured.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed Apache-2.0, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 11 accessibility projects Argusic has installed and timed, agent-desktop was the 4th fastest to reach a running state, and 10 of 11 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 9 minutes after the clone.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
accessibilityaccessibility-apiai-agentsautomationclicomputer-usecomputer-use-agentcomputer-use-agentsdesktop-automationmacosmcprust
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/agent-desktop)Questions
- Does agent-desktop run?
- Yes. agent-desktop runs. Argusic installed and launched it on a clean machine in 9 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test agent-desktop?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit d3dc190afbf3. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test agent-desktop?
- The run that produced this verdict cost $0.23: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does agent-desktop take to install?
- 9 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does agent-desktop need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing agent-desktop?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does agent-desktop compare with the alternatives?
- Of the 11 accessibility projects Argusic has installed and timed, agent-desktop was the 4th fastest to reach a running state, and 10 of 11 reached one at all.
- Where is the evidence for agent-desktop?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.