Zeroshot CLI
zeroshot
Runs coding agents as a graph: one implements, independent ones verify, failures go back for fixing. The agent that wrote the code does not say it works
- 1,960 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Several agents edit code and run commands
- Spends more tokens than one agent
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/zeroshotDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Several agents edit code and run commands
- Spends more tokens than one agent
This is third-party code. Review the repository files before installing.
What it does
Zeroshot builds a graph from coding agents: one agent implements the task, independent agents verify the result, and failures go back for fixing. The idea is that the agent that wrote the code should not claim it works. Installation: `npm install -g @the-open-engine-company/zeroshot`, Node.js 18 or newer is needed. The package installs a verified native binary for Linux x64 and arm64, macOS x64 and arm64, Windows x64 and the Zeroshot skill for Codex and others. There is a Docker image for an isolated target. The project is in Rust, version 10.10.0 from 03.10.2026, MIT license.
Who it is for. For developers who give agents big tasks and want independent verification of the result instead of the agent's word that everything is done.
Good fit when
- You want an agent's result checked by other agents
- You need a graph of several agents with a return for fixing
- You need to run agents in an isolated target through Docker
Not a fit when
- You are not ready for several agents to edit code and run commands
- You need simple work of one agent with no verifiers
Example request
Run a graph: one agent implements the CSV report export, two independent ones check that tests pass and send failures back for fixingLimitations
The agents edit code and run commands, so access is wide: run in an isolated Docker target and on a separate branch. Several agents spend more tokens than one. The package installs a native binary and a skill for Codex and other clients. Version 10.x, so behavior may change between releases.
How to disable. Remove the package with npm uninstall -g @the-open-engine-company/zeroshot, stop the Docker container if you started one and delete the skill from the agent's skills directory.
FAQ
What does Zeroshot do?
It runs coding agents as a graph: one implements, independent ones verify, failures go back for fixing.
How do I install it?
npm install -g @the-open-engine-company/zeroshot, Node.js 18 or newer is needed.
Is there isolation?
Yes, there is a Docker image for an isolated target.
Discussion
No questions or reviews yet.
Related
CLI toolsA multi-agent harness that verifies results by artifacts: stop hooks, an independent judge and an event log for Claude Code, Codex, Cursor and more
A terminal coding agent built on Codex, tuned for low-cost open models like Kimi K3, DeepSeek and Qwen
A skill plus hooks that keep a coding agent on task and cut speculative defenses, unread checksums and scope creep
A self-hosted runtime for coding agents: each crew in its own Linux container, wrapped in a control plane with roles and audit
A self-hosted AgentOS for one workspace: plan-mode chat, private knowledge base, MCP tools, skills, memory and a sandbox
A CLI that detects your project's stack and installs matching agent skills from a vetted registry in one command
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| Zeroshot (this entry) | no | 1,960 | |
| FlowGram | yes | 8,491 | a framework for an AI workflow editor for developers |
| Super Simple Software Factory | no | 969 | agent workflows driven by Python code |
| dzhng/skills | no | 1,293 | a chain from idea to spec to implementation |
README in short
The README describes Zeroshot as launching coding agents as a graph with independent verification, installation through npm and a Docker image. The project is in Rust. MIT license.