Agent-S
Agent S: the state-of-the-art open source computer use agents
Runs with mockssource: GitHubhomepagePythonApache-2.0commit 3aa272d23d29
Python, Apache-2.0 licensed. The project labels itself: agent computer interface, ai agents, computer automation, computer use, computer use agent, cua, grounding and gui agents.
Agent-S runs, with stand-ins for the services it depends on. An Argusic agent installed it in 5 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
gui-agents v0.3.2 installed and its 5 unit tests pass; all S1/S2/S2.5/S3 modules import and engine constructors work correctly.
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 92 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . 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.
- python_requires constraint in setup.py excludes Python 3.12.3, causing pip install to fail1 minute
- Install time
- 5 minutes
- Cold machine to finish
- 8 minutes
- Errors hit and fixed
- 1 hit, 1 fixed with no human help
- How the result was proved
- python -m unittest tests/test_providers.py -v exited 0, 5 passed
- Model tokens used
- 117,591
- Exact commit tested
- 3aa272d23d29
- 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
- Behaviour against real services
- this run was verified against stand-in services only
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
- **Virtual env**: Created `/tmp/agents_venv` since system pip is externally managed.
- **Installed**: `pip install -e ".[dev]"` — all 100+ dependencies including `openai`, `anthropic`,…
- **Tests**: All 5 unit tests in `tests/test_providers.py` pass.
- **Module verification**: All core modules (`LMMAgent`, `OSWorldACI`, `AgentS3`, engine classes fo…
- **CLI caveat**: The `agent_s` console script can't fully launch without `tkinter` (system package…
The `desktop-env` package used in OSWorld evaluation scripts fails to build (`evdev` C extension ca…
exec
/bin/sh -lc "echo '{\"install_succeeded\": true, \"launch_succeeded\": true, \"install_minutes\": 5…
succeeded in 0ms:
{"install_succeeded": true, "launch_succeeded": true, "install_minutes": 5, "errors": [{"msg": "pyt…
tokens used
117,591Replay 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
- Installed in 5 minutes.
- Recovered from all 1 error without a human stepping in, which says the failures are documented well enough to solve.
- Licensed Apache-2.0, as reported by its host.
- Measured 2 times, so the result is not a one-off.
What did not, or is not known
- Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
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
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 2/3 | Runs with mocks | 92.00 | 0.08 | |
| Argusic Runner | 1/3 | Runs with mocks | 92.00 | 0.03 |
Topics (from GitHub)
agent-computer-interfaceai-agentscomputer-automationcomputer-usecomputer-use-agentcuagroundinggui-agentsin-context-reinforcement-learningmemorymllmplanningretrieval-augmented-generation
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/agent-s)Questions
- Does Agent-S run?
- Agent-S runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 5 minutes, hitting 1 error on the way, and recorded the session.
- How did Argusic test Agent-S?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 3aa272d23d29. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test Agent-S?
- The run that produced this verdict cost $0.08: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does Agent-S take to install?
- 5 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-S need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing Agent-S?
- 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for Agent-S?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.