explaingit

coasty-ai/open-cowork

Analysis updated 2026-05-18

39TypeScriptAudience · vibe coderComplexity · 3/5LicenseSetup · easy

TLDR

An open-source AI coworker that watches a screen and completes computer tasks step by step, pausing for your approval before anything billable happens.

Mindmap

mindmap
  root((open-cowork))
    What it does
      AI coworker for screen tasks
      Live streaming and approval
      Cost caps enforced
    Tech stack
      TypeScript
      Node.js
      Electron
    Use cases
      Delegate a computer task
      Try free demo mode
      Bring your own LLM
    Audience
      Vibe coders
      Automation builders

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Delegate a computer task in plain text and watch an AI agent carry it out on your desktop, a cloud VM, or in a browser.

USE CASE 2

Run the tool in free demo mode with a mock server before connecting a real API key.

USE CASE 3

Bring your own model through OpenRouter, OpenAI, or a local Ollama or LM Studio instance instead of the default provider.

USE CASE 4

Build multi-step workflows with branching, retries, and human approval checkpoints for more complex automations.

What is it built with?

TypeScriptNode.jspnpmElectronReact Native

How does it compare?

coasty-ai/open-coworkjavlonbek1233/-immersive-ui-1javlonbek1233/-labx-virtual-science
Stars393939
LanguageTypeScriptTypeScriptTypeScript
Setup difficultyeasyeasymoderate
Complexity3/51/52/5
Audiencevibe codervibe coderdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 5min

Runs in a zero-cost demo mode out of the box, a real Coasty or BYOK key is only needed for live automation.

Use freely for any purpose, including commercial use, as long as you keep the copyright notice.

In plain English

open-cowork is an open-source tool that lets you describe a computer task in plain text and then watch an AI agent carry it out on screen, step by step. You can point it at your own desktop, a cloud virtual machine, or a browser window. As the agent works, you see a live stream of what it is doing, and it pauses to ask for your approval before taking actions that matter. The project is described as an open-source alternative to Claude Co-Work, built using the Coasty Computer Use API by default. That API handles the part where an AI model looks at a screen and decides what to click or type next. If you do not want to use Coasty, you can switch to any other provider in settings: OpenRouter, OpenAI, or a locally running model through Ollama or LM Studio. This is the BYOK (bring your own key, or bring your own LLM) mode. Setup takes one command after cloning the repo: pnpm desktop. That starts the backend, opens the web interface, and launches a desktop app that can control your actual mouse and keyboard. With no API key configured it runs in demo mode using a mock server and a sandbox key, so nothing is billed and no account is needed to try it out. For more involved use cases, the tool includes a workflow builder where you write tasks as a JSON description that supports branching, loops, retries, parallel steps, and human approval checkpoints. Each run shows a cost estimate upfront and requires you to confirm before anything billable starts. There are also hard budget caps enforced on the server side. You can monitor and manage runs from a web dashboard, a desktop window, or a mobile app. If a run pauses waiting for your approval, a notification appears on your phone so you can unblock it from wherever you are. The event log is durable and can replay from the beginning, which helps when you want to review what the agent did. The codebase is a TypeScript monorepo (Node 22+, pnpm) covering the backend API, the web frontend, the Electron desktop app, and a React Native mobile app.

Copy-paste prompts

Prompt 1
Clone this repo and run pnpm desktop so I can try the demo mode with no API key.
Prompt 2
Help me switch this open-cowork setup to use my own LLM through OpenRouter instead of Coasty.
Prompt 3
Explain how the approval flow works when a task needs a human to confirm before continuing.
Prompt 4
Show me how to write a workflow JSON with branching and retries using this project's DSL.
Prompt 5
Set up webhooks for this project so I get live status updates without polling.

Frequently asked questions

What is open-cowork?

An open-source AI coworker that watches a screen and completes computer tasks step by step, pausing for your approval before anything billable happens.

What language is open-cowork written in?

Mainly TypeScript. The stack also includes TypeScript, Node.js, pnpm.

What license does open-cowork use?

Use freely for any purpose, including commercial use, as long as you keep the copyright notice.

How hard is open-cowork to set up?

Setup difficulty is rated easy, with roughly 5min to a first successful run.

Who is open-cowork for?

Mainly vibe coder.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.