agent-sandbox

Secure local dev environment for collaboration with AI coding agents

Runssource: GitHubPythonMITcommit c5b65e7cbd8f

Python, MIT licensed. The project labels itself: agent harness, agent sandbox, agents and coding agents.

agent-sandbox runs. An Argusic agent installed it in 8 minutes and hit 2 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

The agentbox CLI binary is built and functional at dist/agentbox-linux-amd64, all 5 Go test packages pass, and all 195 proxy Python tests pass.

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

verdict
Runs
Argusic Score
100 / 100
cost of the verifying run
$0.07 (measured)
recorded runs
1
last tested
stars
214
forks
19
open issues
17
watchers
2
size
3 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 agent-sandbox, 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.

  • Go 1.26.1 compiler not found in container3 minutes
  • PyYAML module missing for proxy Python tests1 minute
Install time
8 minutes
Cold machine to finish
4 minutes
Errors hit and fixed
2 hit, 2 fixed with no human help
How the result was proved
go test ./... exited 0 with 5 packages passing (cli, docker, runtime, scaffold, version); python3 -m unittest discover -s images/proxy/tests -p 'test_*.py' exited 0 with 195 tests passing (17 skipped); dist/agentbox-linux-amd64 version printed Agent Sandbox v0.0.0
Model tokens used
56,662
Exact commit tested
c5b65e7cbd8f
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.

commit: c5b65e7cbd8f5b3bbf4e3ea40900c0014eedfa04
commit-time: 2026-09-26T23:00:07Z
source: buildinfo
go: go1.26.1
===
Manage secure local agent sandboxes
Usage:
  agentbox [command]
Here's my final report.
tokens used
56,662
Here's my final 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

  • Reached a running state on a clean machine, with the session recorded.
  • Recovered from all 2 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 MIT, as reported by its host.

What did not, or is not known

  • Took 8 minutes to install, slower than the median of the 10 comparable projects Argusic has measured.
  • Measured once so far. A single attempt shows it can run, not that it runs reliably.

Of the 11 agent-harness projects Argusic has installed and timed, agent-sandbox was the 7th fastest to reach a running state, and 9 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 8 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

tested byattemptoutcomescorecost USDwhen
Argusic Runner1/3Runs100.000.07

Topics (from GitHub)

agent-harnessagent-sandboxagentscoding-agents

Embed the badge

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

[![Tested by Argusic](https://argusic.com/badge/agent-sandbox.svg)](https://argusic.com/subject/agent-sandbox)

Questions

Does agent-sandbox run?
Yes. agent-sandbox runs. Argusic installed and launched it on a clean machine in 8 minutes, hitting 2 errors on the way, and recorded the session.
How did Argusic test agent-sandbox?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit c5b65e7cbd8f. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test agent-sandbox?
The run that produced this verdict cost $0.07: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does agent-sandbox take to install?
8 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-sandbox need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing agent-sandbox?
2 things broke in the recorded run, and 2 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does agent-sandbox compare with the alternatives?
Of the 11 agent-harness projects Argusic has installed and timed, agent-sandbox was the 7th fastest to reach a running state, and 9 of 11 reached one at all.
Where is the evidence for agent-sandbox?
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.

Discussion