SWE-AF

Autonomous software engineering fleet of AI agents for production-grade PRs on AgentField: plan, code, test, and ship.

Runs with mockssource: GitHubhomepageGoApache-2.0commit 17d3160f20a2

Go, Apache-2.0 licensed. The project labels itself: agent factory, agentfield, agentic ai, ai agents, ai factory, autonomous agents, claude code and code generation.

SWE-AF runs, with stand-ins for the services it depends on. An Argusic agent installed it in 0.3 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

The swe-af package is installed in the venv, all 1317 tests pass, and python -m swe_af starts a uvicorn server on port 8001 (reports degraded mode because no AgentField control plane is running on localhost:9999).

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 56 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs with mocks
Argusic Score
56 / 100
cost of the verifying run
$0.12 (measured)
recorded runs
2
last tested
stars
1,136
forks
187
open issues
25
watchers
9
size
128 MB
created
last push

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.

Time lost to each failure while testing SWE-AF, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

  • test_default_runtime_and_provider, test_empty_env_uses_claude_code, test_unset_env_uses_claude_code, test_default_resolution failed from same env-var interference1.5 minutes
  • test_ac_3_fastbuildconfig_defaults failed because OPENROUTER_API_KEY in container env changed default runtime from claude_code to open_code0.5 minutes
  • test_build_config_no_args (multi_repo_schemas) same env interference0.5 minutes
  • test_ac_25 (multi_repo_integration) same env interference in child subprocess0.5 minutes
Install time
under a minute
Cold machine to finish
14 minutes
Errors hit and fixed
4 hit, 4 fixed with no human help
How the result was proved
pytest tests/ exited 0 with 1317 passed, 1 skipped
Model tokens used
253,958
Exact commit tested
17d3160f20a2
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.

ℹ️ Reconnecting to memory events (attempt 3/5) in 4.0s...
❌ Failed to connect to memory events: Multiple exceptions: [Errno 111] Connect call failed ('127.0.…
ℹ️ Reconnecting to memory events (attempt 4/5) in 8.0s...
❌ Failed to connect to memory events: Multiple exceptions: [Errno 111] Connect call failed ('127.0.…
ℹ️ Reconnecting to memory events (attempt 5/5) in 16.0s...
INFO:     Shutting down
Now let me compile the final report:
**ARGUSIC_RESULT: {"install_succeeded": true, "launch_succeeded": true, "install_minutes": 0.28, "e…
tokens used
253,958
Now let me compile the final report:
**ARGUSIC_RESULT: {"install_succeeded": true, "launch_succeeded": true, "install_minutes": 0.28, "e…

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 0.3 minutes.
  • Recovered from all 4 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.
  • 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 byattemptoutcomescorecost USDwhen
Argusic Runner2/3Runs with mocks92.000.12
Argusic Runner1/3Did not run20.000.19

Topics (from GitHub)

agent-factoryagentfieldagentic-aiai-agentsai-factoryautonomous-agentsclaude-codecode-generationcodexcoding-agentdeveloper-toolsgemini-clillmmulti-agentopenroutersoftware-engineeringsoftware-factory

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/SWE-AF.svg)](https://argusic.com/subject/swe-af)

Questions

Does SWE-AF run?
SWE-AF runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 0 minutes, hitting 4 errors on the way, and recorded the session.
How did Argusic test SWE-AF?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit 17d3160f20a2. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test SWE-AF?
The run that produced this verdict cost $0.12: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does SWE-AF take to install?
0.3 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 SWE-AF need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing SWE-AF?
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.
Where is the evidence for SWE-AF?
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.

Discussion