# Browser & Computer Use

> Agents that operate web browsers and computers to automate real-world tasks.

Browser and computer-use agents operate a real browser or desktop the way a person would — clicking, typing, scrolling, and navigating — to automate tasks that have no API. That makes them powerful for scraping, testing, form-filling, and cross-app workflows, and also the hardest kind of agent to make reliable, because real websites change, fight bots, and break automation in subtle ways. Choose based on success rate on your actual target sites, speed, and how it handles the messy parts: logins, multi-factor prompts, CAPTCHAs, pop-ups, and shifting layouts. A key split is where it runs — a local automation you fully control versus a managed cloud environment that handles browser fleets, proxies, and session persistence for you, usually more robustly but as metered usage. Look at how it grounds actions — raw pixels, the accessibility tree, or the DOM — since that drives both accuracy and cost, and whether you can record, replay, and debug a run when it fails. Pricing is commonly per-action or per-hour of compute, which adds up quickly at scale. Treat reliability claims skeptically and pilot on your real workflows before depending on one: the difference between eighty and ninety-nine percent success is the difference between a demo and something you can schedule unattended overnight.

20 tools indexed. Sponsored placements are labeled and never ranked; open-source ordering reflects real GitHub adoption. Browse on the web: https://agentsearchengine.app/category/browser-agents

## Commercial products (alphabetical — stars don't measure commercial adoption)
- [Browser Use Cloud](https://agentsearchengine.app/agents/browser-use-cloud) — Cloud API to make websites accessible to AI agents. · Freemium · Platform
- [ChatGPT Agent](https://agentsearchengine.app/agents/openai-operator) — OpenAI's agent mode that browses, clicks, and completes tasks on the web. · Paid product · Agent

## Open-source index (ranked by adoption)
1. [UI-TARS-desktop](https://agentsearchengine.app/agents/ui-tars-desktop) — The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra · Open source · Agent · 38k★ · TypeScript · Apache-2.0 · active 1 day ago · best for vision-driven gui automation across desktop, browser, and terminal
2. [page-agent](https://agentsearchengine.app/agents/page-agent) — JavaScript in-page GUI agent. Control web interfaces with natural language. · Open source · Agent · 28k★ · TypeScript · MIT · active 7 days ago · best for embedding an ai agent into your own web page with a single script
3. [skyvern](https://agentsearchengine.app/agents/skyvern) — Automate browser based workflows with AI · Open source · Agent · 23k★ · Python · AGPL-3.0 · active today · best for automating browser workflows with vision-based agents you can self-host
4. [midscene](https://agentsearchengine.app/agents/midscene) — AI-powered, vision-driven UI automation for every platform. · Open source · Agent · 14k★ · TypeScript · MIT · active today · best for writing ui tests in natural language across web, ios, and android
5. [nanobrowser](https://agentsearchengine.app/agents/nanobrowser) — Open-Source Chrome extension for AI-powered web automation. Run multi-agent workflows using your own LLM API key. Alternative to OpenAI Operator. · Open source · Agent · 14k★ · TypeScript · Apache-2.0 · active 8 months ago · best for a free in-browser alternative to openai operator using your own llm keys
6. [Agent-S](https://agentsearchengine.app/agents/agent-s) — Agent S: an open agentic framework that uses computers like a human · Open source · Framework · 12k★ · Python · Apache-2.0 · active 2 days ago · best for benchmark-leading computer-use automation driven by screenshots
7. [robotgo](https://agentsearchengine.app/agents/robotgo) — RobotGo, Go Native cross-platform RPA, GUI automation, Auto test and Computer use @vcaesar · Open source · SDK / library · 11k★ · Go · Apache-2.0 · active 5 days ago · best for go-native desktop automation for mouse, keyboard, screen, and global events
8. [llm-scraper](https://agentsearchengine.app/agents/llm-scraper) — Turn any webpage into structured data using LLMs · Open source · SDK / library · 6.9k★ · TypeScript · MIT · active 1 day ago
9. [fara](https://agentsearchengine.app/agents/fara) — Fara-7B: An Efficient Agentic Model for Computer Use · Open source · Agent · 6.1k★ · Python · MIT · active 12 days ago
10. [TuriX-CUA](https://agentsearchengine.app/agents/turix-cua) — This is the official website for TuriX Computer-use-Agent · Open source · Agent · 3.1k★ · Python · MIT · active 4 days ago
11. [open-computer-use](https://agentsearchengine.app/agents/open-computer-use) — AI computer use powered by open source LLMs and E2B Desktop Sandbox · Open source · Agent · 2.2k★ · Python · Apache-2.0 · active 25 days ago
12. [RPA](https://agentsearchengine.app/agents/rpa) — Ui.Vision Open-Source RPA Software with Computer Vision, OCR, Anthropic Computer Use/LLM. Selenium IDE import/export. · Open source · Agent · 2.0k★ · JavaScript · active 2 days ago
13. [ShowUI](https://agentsearchengine.app/agents/showui) — [CVPR 2025] Open-source, End-to-end, Vision-Language-Action model for GUI Agent & Computer Use. · Open source · Agent · 1.9k★ · Python · Apache-2.0 · active 3 months ago
14. [OpenAdapt](https://agentsearchengine.app/agents/openadapt) — Open Source Generative Process Automation (i.e. Generative RPA). AI-First Process Automation with Large ([Language (LLMs) / Action (LAMs) / Multimodal (LMMs)] / Visual Language (VLMs)) Models · Open source · Agent · 1.7k★ · Python · MIT · active 1 day ago
15. [ghost-os](https://agentsearchengine.app/agents/ghost-os) — Full computer-use for AI agents. Self-learning workflows. Native macOS. No screenshots required. · Open source · Agent · 1.6k★ · Swift · MIT · active 4 months ago
16. [parchi](https://agentsearchengine.app/agents/parchi) — Your AI friend right in your browser · Open source · Agent · 541★ · TypeScript · MIT · active 3 months ago
17. [openbrowser-ai](https://agentsearchengine.app/agents/openbrowser-ai) — OpenBrowser is a framework for intelligent browser automation. · Open source · Framework · 237★ · Python · MIT · active 27 days ago
18. [BrowserPilot](https://agentsearchengine.app/agents/browserpilot) — Open‑source alternative to Perplexity Comet, director.ai and firecrawl combined · Open source · Agent · 173★ · Python · MIT · active 8 days ago

## Frequently asked
**What is a browser agent?**
A browser agent is an AI that controls a web browser (and sometimes a full computer) to complete tasks visually — useful when a site has no API.

**How do I choose a computer-use agent?**
Test it on your actual target sites: success rate, speed, session/login handling, and whether you can run it in a controlled cloud environment for reliability and safety.

**Are browser agents reliable?**
They've improved a lot but still struggle with anti-bot measures and frequently-changing UIs. Managed cloud offerings tend to be more robust than raw local automation.

---
Source: [Agent Search Engine](https://agentsearchengine.app). Full directory in one file: https://agentsearchengine.app/llms-full.txt