zeroshot

Build your own open-source software factory.

Turn a task into a reviewed pull request through a workflow you control. Choose the agents and models, define what must pass, and decide where rejected work goes for repair.

Start with the built-in software-change graph: implementation → independent review → repair → optional delivery. Or design your own topology. Add security reviewers, test agents, and parallel checks, then save the workflow as a profile and reuse it.

The agent that writes the code cannot approve its own work. The engine enforces the graph's review gates and retry limits. Think of it as CI for coding agents, starting before the code is written. zeroshot runs Claude Code, Codex, or Copilot as the worker and reviewers, with a separate harness and provider available for each agent.

The CLI is free and MIT licensed. Run locally, self-host, or use zeroshot Cloud.

Workflow illustration: both reviews run again after a repair. Git delivery is optional. Static diagram

We want to explore how much task decomposition and validation loops can narrow the gap between open models and closed frontier systems. The graph, review rules, and repair paths are open for anyone to inspect, change, and test on real work.

▶ Video: a single agent vs. zeroshot (scripted)

Questions, ideas, or a run worth showing? Join the zeroshot community on Discord.

Get started

Prefer Cloud? Cloud setup →

Run locally

npm install -g @the-open-engine-company/zeroshot

The installer requires Node.js 18 or newer and installs a verified native binary for Linux x64/arm64, macOS x64/arm64, or Windows x64. It also installs one zeroshot skill for Codex, GitHub Copilot, and Claude Code at user scope.

For local execution, install and sign in to Codex, Claude Code, or GitHub Copilot. Local runs can reuse the harness's existing login, including subscription-backed sessions. See the installation guide for harness prerequisites.

Run your first task

This example uses Codex and keeps delivery local. The worker edits the current Git worktree, so start in a clean worktree intended for the task.

Create input.json, replacing the example task with a change appropriate for your repository:

{
  "task": "Add JSON output to the status command and cover it with focused tests."
}

Create runtime.json. Replace YOUR_MODEL_ID with a model identifier supported by your installed Codex CLI:

{
  "harness": "codex",
  "provider": "openai",
  "model": "YOUR_MODEL_ID",
  "effort": "high"
}

Start the run. Add --validate-only to check the graph, runtime configuration, and input first:

zeroshot run \
  --title "Add JSON status output" \
  --template software-change \
  --input input.json \
  --uniform-runtime-config runtime.json

Open another terminal to inspect the run:

zeroshot ui

Visit http://127.0.0.1:4173/ui/ and open Runs. Stopping the UI server leaves active runs running. The CLI also provides zeroshot list, zeroshot status RUN_ID, and zeroshot logs RUN_ID.

For other harnesses and providers, see Runtimes and connections.