AgentScout / Tasks / Browsing the web / Computer and desktop control

Computer and desktop control tools for web browsing agents

Agents that see the screen and use the mouse and keyboard, so they can work in desktop apps as well as the browser. Of the 104 tools on our browsing the web page, 24 fit this narrower job. By pricing: 18 open source, 4 free or freemium, 1 paid, usage-based or enterprise, 1 pricing on the maker's site. The most-starred open repos here are UI-TARS-desktop (39,297 stars), Cua (29,966 stars), Midscene.js (15,311 stars). Most come from the browser/computer use (19), personal agents (2), productivity (1) categories.

Updated

UI-TARS-desktop logo

UI-TARS-desktop

browser/computer useOpen source

ByteDance's open-source multimodal agent stack, including a desktop app and Agent TARS for controlling computers and browsers.

Added

Cua logo

Cua

browser/computer useOpen source + cloud

Open-source infrastructure for computer-use agents: sandboxes, drivers across operating systems and benchmarks.

Added

Midscene.js logo

Midscene.js

Newbrowser/computer useOpen source

Vision-driven UI automation and E2E testing framework where a GUI agent carries out natural-language steps on web, mobile and desktop apps.

Added

RobotGo logo

RobotGo

Newbrowser/computer useOpen source

Go library for cross-platform desktop automation and AI computer use that controls mouse, keyboard, screen, windows and processes.

Added

UFO logo

UFO

Newbrowser/computer useOpen source

Microsoft's desktop agent framework that operates Windows apps and orchestrates multi-device tasks as task graphs.

Added

Windows-MCP logo

Windows-MCP

Newbrowser/computer useOpen source

Lightweight MCP server for computer use on Windows that lets agents navigate files, control apps and interact with the UI.

Added

Peekaboo logo

Peekaboo

Newbrowser/computer useOpen source

macOS CLI, menu-bar app and MCP server for screen capture, accessibility inspection and native UI automation by AI agents.

Added

AstronRPA logo

AstronRPA

Newbrowser/computer useOpen source

iFlytek's open-source enterprise RPA desktop app with a low-code visual designer whose workflows can be called by AI agents.

Added

Cumora logo

Cumora

NewproductivityOpen source

Cumora is a team chat app where AI agents are members of the workspace rather than a chat box you prompt. Each agent gets a persona, its own private memory and files, a status, and a schedule on which it wakes up, reads the room and decides whether to message someone, post an idea or pull a small group together. Agents can message each other, and you can read those side conversations. People join by email or link, agents can be grouped into teams and given projects, and the same live workspace runs on desktop, in the browser and on a phone. It is free during an invite-only preview.

Added

Handsy logo

Handsy

Newbrowser/computer useOpen source

Lets an AI model use a real iPhone through a $7 ESP32 board that acts as a Bluetooth keyboard and mouse, tapping and typing through apps with no jailbreak and no computer attached.

Added

TuriX CUA logo

TuriX CUA

browser/computer useOpen source

Computer-use agent that operates desktop apps.

Added

Skales logo

Skales

Newpersonal agentsFreemium

Personal AI agent for desktop and mobile that codes, automates the desktop and browser and runs scheduled tasks with local or cloud models.

Added

Ui.Vision RPA logo

Ui.Vision RPA

Newbrowser/computer useOpen source

Browser and desktop automation extension with OCR and image recognition, plus an MCP server so AI agents can create and run macros.

Added

LCU logo

LCU

Newbrowser/computer useOpen source

Brings the computer-use runtime from the ChatGPT desktop app to Claude Code, Codex CLI and Pi, so any agent harness can see and control your desktop without a Codex sign-in.

Added

OpenAdapt logo

OpenAdapt

Newbrowser/computer useOpen source

Execution layer for computer-use agents that turns a demonstrated task into a local program that repeats the clicks without model calls and stops on mismatches.

Added

Terminator logo

Terminator

Newbrowser/computer useOpen source

Desktop automation library and MCP server that lets AI assistants control any Windows app through the accessibility tree, like Playwright for the desktop.

Added

AgentBay logo

AgentBay

Newsandboxes & code executionUsage-based

Alibaba Cloud's cloud sandbox service for agents, offering ready-made browser, Linux and Windows desktop, mobile and code environments that agents control through an SDK or MCP.

Added

OpenGCUA logo

OpenGCUA

Newbrowser/computer useOpen source

Open stack from Tongyi Lab's MAI-UI team for generalist computer-use agents that combine GUI clicks, accessibility trees, terminal commands, APIs and code: an OSWorld harness, a macOS real-device harness, a cross-device SDK and published task runs of eight frontier models.

Added

Dodaios logo

Dodaios

Newpersonal agentsOpen source

Open-source desktop agent OS for Windows, macOS and Linux (formerly Personal Jarvis): talk to it and it runs your coding CLIs side by side, your browser and your apps, with offline Whisper speech, a local wiki memory and support for Ollama, Claude or ChatGPT subscriptions.

Added

Orke logo

Orke

NewcodingFree

BOMTOBE's free desktop workspace for Claude Code, Codex and other coding agents: terminal sessions, an in-app browser the agent can use, a Git review window and scheduled AI employees.

Added

AGI-0 by AGI, Inc. logo

AGI-0 by AGI, Inc.

Newbrowser/computer useSee site

AGI, Inc.'s small action model and Android agent that sees the screen and acts across any app without APIs.

Added

Anchor Action Index logo

Anchor Action Index

browser/computer useFreemium

Shared memory of web workflows for Anchor's cloud browsers: computer-use agents replay proven paths as deterministic code and the model only steps in where a page has changed, exposed as tools on Anchor's MCP server and REST API.

Added

Arro logo

Arro

browser/computer useFreemium

Arro, a desktop assistant that guides you through tasks on your computer and repeats them for you.

Added

Holo4 (H Company) logo

Holo4 (H Company)

browser/computer useOpen weights

Open-weight computer-use models (27B and 35B-A3B) released Oct 1.

Added

GitHub repos (20)

RepoGitHub starsLicenceLast push
UI-TARS-desktop ↗39,297Apache License 2.02026-10-09
Cua ↗29,966MIT2026-10-21
Midscene.js ↗15,311MIT2026-10-21
RobotGo ↗10,871Apache License 2.02026-10-07
UFO ↗10,056MIT2026-10-14
Windows-MCP ↗9,404MIT2026-10-20
Peekaboo ↗5,335—2026-10-21
AstronRPA ↗5,276—2026-10-20
Cumora ↗4,005——
Handsy ↗3,377—2026-10-19
TuriX CUA ↗3,171MIT2026-09-14
Skales ↗2,196—2026-10-19
Ui.Vision RPA ↗2,083—2026-09-21
LCU ↗1,913—2026-10-21
OpenAdapt ↗1,782—2026-10-21

Top 15 of 20 by stars. Stars and licence come from each repo, checked when the listing was last verified.

FAQ

What are the best AI tools for browsing the web focused on computer and desktop control?

We list 24. The first three by GitHub stars and featured status are UI-TARS-desktop: ByteDance's open-source multimodal agent stack, including a desktop app and Agent TARS for controlling computers and browsers.; Cua: Open-source infrastructure for computer-use agents: sandboxes, drivers across operating systems and benchmarks.; Midscene.js: Vision-driven UI automation and E2E testing framework where a GUI agent carries out natural-language steps on web, mobile and desktop apps.

Are there free or open-source options?

Yes: 18 open source, 4 free or freemium, 1 paid, usage-based or enterprise, 1 pricing on the maker's site. Pricing is as each maker states it.

More from browsing the web

Tools are grouped by what each says it does. Missing one? Submit it free.

Get the best new AI agents, weekly

One short email with new tools, launches and what's worth trying. Free, unsubscribe anytime.