zeroshot
Runs coding agents as a graph: one agent implements, independent agents review, failures go to repair, and nothing ships until the checks pass. Works with Codex, Claude Code and Copilot.
Runssource: GitHubhomepageRustMITcommit 63d4832da7f2
Rust, MIT licensed. The project labels itself: agent loops, agent orchestration, agentic workflow, ai agent, ai agents, autonomous agents, claude and claude code.
zeroshot runs. An Argusic agent installed it in 28 minutes and hit 4 errors on a clean machine with no GPU, and the whole session was recorded.
Zeroshot installs via npm and builds from source; the CLI version, template list, UI server (200 on /ui/), and run validation all work, with 1607/1608 Rust tests passing in this container.
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 95 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 fetch_waits_for_automatic_maintenance fails: git 2.43 runs maintenance without --no-detach; repo requires git >= 2.47 and fixture hardcodes /usr/bin/git10 minutes
- cargo link died with Bus error / No space left on device (36G debug target on 40G root)8 minutes
- rustc/cargo missing from container5 minutes
- npm global install failed EACCES on /usr/local prefix2 minutes
- Install time
- 28 minutes
- Cold machine to finish
- 25 minutes
- Errors hit and fixed
- 4 hit, 3 fixed with no human help
- How the result was proved
- cargo test --workspace returned 1607 passed/1 failed; npm test 64 pass; ui test 348 pass; zeroshot ui served GET /ui/ 200 and GET /ui/api/runs 200; zeroshot run --validate-only returned {"valid":true}
- Model tokens used
- 770,329
- Exact commit tested
- 63d4832da7f2
- 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.
- Installed the Rust toolchain (1.97.0 per `rust-toolchain.toml`) since none was present.
- `cargo build --workspace` succeeded. Full `cargo test --workspace`: 1607 passed, 1 failed, 8 igno…
- Node tooling: `npm test` 64/64 passed; ESLint clean; `distribution:check` passed; `protocol:check…
- UI (Node 24.21.0, installed locally since system Node is 18): `npm --prefix ui ci`, `npm --prefix…
- Source-built binary with `--features ui` launched: `zeroshot version` → `zeroshot 0.1.0` (develop…
- Real pipeline check: `zeroshot run --template software-change --validate-only` returned `{"valid"…
**Errors found and fixed**
- npm global install EACCES on `/usr/local` (unwritable prefix) → set user prefix, reinstalled.
- `rustc`/`cargo` missing → installed via rustup (stable + 1.97.0).
- Linker `Bus error`/`No space left on device` during test/build (debug `target/` hit the 40G conta…
- One unfixable environment gap: `native_v2_hosting::repository::tests::fetch_waits_for_automatic_m…
I did not start an agent run (AGENTS.md forbids `zeroshot run` unless explicitly requested), so end…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.
- Installed in 28 minutes.
- 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
- Hit 4 errors during setup, 1 of which were not solved.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
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
Topics (from GitHub)
agent-loopsagent-orchestrationagentic-workflowai-agentai-agentsautonomous-agentsclaudeclaude-codeclicode-reviewcodexcoding-assistantdeveloper-toolsgithub-copilotloop-engineeringmeta-harnessmulti-agentvibecoding
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/zeroshot)Questions
- Does zeroshot run?
- Yes. zeroshot runs. Argusic installed and launched it on a clean machine in 28 minutes, hitting 4 errors on the way, and recorded the session.
- How did Argusic test zeroshot?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 63d4832da7f2. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test zeroshot?
- The run that produced this verdict cost $0.22: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does zeroshot take to install?
- 28 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 zeroshot need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing zeroshot?
- 4 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 zeroshot?
- 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.