cuga-agent
CUGA is an open-source generalist agent harness for the enterprise, supporting complex task execution on web and APIs, OpenAPI/MCP integrations, composable architecture, reasoning modes, and policy-aware features.
Runs with mockssource: GitHubhomepagePythonApache-2.0commit bd31b638b1d8
Python, Apache-2.0 licensed. The project labels itself: coding agent, computer use, enterprise, generalist agent, guardrails, harness, knowledge and mcp.
cuga-agent runs, with stand-ins for the services it depends on. An Argusic agent installed it in 1.5 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
CUGA 0.4.1 installs successfully with uv, imports cleanly, and 2358 of 2362 unit tests pass with a dummy OPENAI_API_KEY; 3 test_sdk_citations failures are ordering-dependent and pass when run in isolation
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.
- test_e2e_server_stack.py: subprocess.Popen(['python', ...]) fails because 'python' is not on PATH in this container0.5 minutes
- evolve extra was missing (altk_evolve module not found), causing 7 evolve retention tests and 1 evolve bundled hooks test to fail0.3 minutes
- test_agent_policy_isolation.py requires OPENAI_API_KEY env var; fails with credential error when absent0.2 minutes
- Install time
- 2 minutes
- Cold machine to finish
- 29 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- pytest tests/unit/ -q returned '2358 passed, 1 skipped' with only 3 pre-existing ordering-dependent failures; 'cuga --help' exited with usage message; 'import cuga' succeeded; app server imported with 54 routes
- Model tokens used
- 888,668
- Exact commit tested
- bd31b638b1d8
- 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.
process = subprocess.Popen([sys.executable, "-c", "raise SystemExit(3)"])
process.wait(timeout=5)
with pytest.raises(RuntimeError, match="died with return code"):
Now let me produce the final result
exec
/bin/sh -lc 'export PATH="$HOME/.local/bin:$PATH" && cd /work/repo && OPENAI_API_KEY=sk-test-fake .…
succeeded in 8468ms:
ERROR: not found: /work/repo/tests/unit/acp/test_outbound_contract.py::test_diagnose_returns_stderr…
(no match in any of [<Module test_outbound_contract.py>])
no tests ran in 0.03s
tokens used
888,668Replay 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 1.5 minutes.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Licensed Apache-2.0, as reported by its host.
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.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
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 | 1/3 | Runs with mocks | 92.00 | 0.14 |
Topics (from GitHub)
coding-agentcomputer-useenterprisegeneralist-agentguardrailsharnessknowledgemcppoliciessandbox
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/cuga-agent)Questions
- Does cuga-agent run?
- cuga-agent runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 2 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test cuga-agent?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit bd31b638b1d8. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test cuga-agent?
- The run that produced this verdict cost $0.14: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does cuga-agent take to install?
- 1.5 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 cuga-agent need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing cuga-agent?
- 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.
- Where is the evidence for cuga-agent?
- 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.