AWorld
Search, understand, reproduce, and improve an idea with ease
Runs with mockssource: GitHubhomepagePythonMITcommit 631f67f54b68
Python, MIT licensed. The project labels itself: agent framework, agent learning, agent runtime, browsecomp, environment, gaia, mcp and rl training.
AWorld runs, with stand-ins for the services it depends on. An Argusic agent installed it in 2 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
AWorld and AWorld-CLI packages install and build successfully. All 170 ACP unit tests pass with memory bootstrap gracefully degrading when no LLM API key is configured. The ACP self-test bridge validates 15 phase-1 protocol cases end-to-end. The aworld and aworld-cli entry points respond to --help 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 84.5 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.
- ACP server crashed on startup: _bootstrap_acp_memory() needed OpenAI API key even in self-test mode30 minutes
- Test test_validate_stdio_host_command_accepts_topology_template fails: template command uses 'python' but only 'python3' exists on system10 minutes
- ACP test_acp_stdio_harness_can_launch_local_server_and_initialize fails: runs without AWORLD_ACP_SELF_TEST_BRIDGE, still hits memory init crash5 minutes
- ACP test_acp_self_test_reports_required_phase1_case_ids fails: subprocess returns exit code 1 because aworld_cli module not installed2 minutes
- Install time
- 2 minutes
- Cold machine to finish
- 51 minutes
- Errors hit and fixed
- 4 hit, 4 fixed with no human help
- How the result was proved
- pytest tests/acp/ -- 170 passed; pytest aworld-cli/tests/ -- 239 passed; aworld_cli.main --no-banner acp validate-stdio-host --topology same-host returned ok=true (15/15 cases passed); aworld --help and aworld-cli --help both report valid usage
- Model tokens used
- 393,403
- Exact commit tested
- 631f67f54b68
- 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.
Here's the report: 2026-11-18T05:43:45.970519Z ERROR codex_core::session: failed to record rollout items: thread 01a1e… 2026-11-18T05:43:45.970562Z ERROR codex_core::session: failed to record rollout items: thread 01a1e… 2026-11-18T05:43:45.970572Z ERROR codex_core::session: failed to record rollout items: thread 01a1e… tokens used 393,403 Let me compile the final report now. I have strong evidence from multiple test runs: - Install: succeeded - ACP tests: 170/170 passed - aworld-cli tests: 239/239 passed - End-to-end ACP self-test validation: 15/15 cases passed Here's the report:
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
- Installed in 2 minutes, faster than the median of the 63 comparable projects Argusic has measured.
- Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
- Licensed MIT, 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.
Of the 64 agent-framework projects Argusic has installed and timed, AWorld was the 21st fastest to reach a running state, and 51 of 64 reached one at all.
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 | 0.49 | |
| Argusic Runner | 1/3 | Runs with mocks | 77.00 | 0.20 |
Topics (from GitHub)
agent-frameworkagent-learningagent-runtimebrowsecompenvironmentgaiamcprl-trainingworld-modelxbench
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/aworld)Questions
- Does AWorld run?
- AWorld runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 2 minutes, hitting 4 errors on the way, and recorded the session.
- How did Argusic test AWorld?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 631f67f54b68. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test AWorld?
- The run that produced this verdict cost $0.49: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does AWorld take to install?
- 2 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 AWorld need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing AWorld?
- 4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does AWorld compare with the alternatives?
- Of the 64 agent-framework projects Argusic has installed and timed, AWorld was the 21st fastest to reach a running state, and 51 of 64 reached one at all.
- Where is the evidence for AWorld?
- 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.