cua

Automate browser tasks across Steel, Browserbase, and Browser Use cloud providers.

9|5|Updated Jan 22, 2026

Search Tags:#playwright #puppeteer #selenium #web-scraping #browser-automation #captcha #cloud-browsers

Ready to run

npx skills add https://github.com/x402agent/SolanaOS --skill cua

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: cua
Source: https://github.com/x402agent/SolanaOS/tree/main/skills/cua
Command: npx skills add https://github.com/x402agent/SolanaOS --skill cua

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autonomous cloud browser control that orchestrates CDP-based sessions across Steel, Browser Use, and Browserbase to automate web tasks without manual browser interaction.

Core Features & Use Cases

Multi-provider cloud sessions: Create and manage headful browser sessions across Steel, Browser Use, and Browserbase for automation, testing, and data collection.
LLM-assisted navigation and automation: Vision-driven decision making and Playwright/Puppeteer execution guided by a large language model.
Use Case: Automate form filling, dynamic site data extraction, and automated testing across cloud browsers with live view and session replay.

Quick Start

Use cua to initialize a cloud browser session and begin autonomous web automation.

Frequently Asked Questions about cua

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate web scraping with cloud browsers instead of local Playwright scripts?▼

This Skill initializes headful cloud browser sessions via CDP connections across Steel, Browserbase, and Browser Use, enabling autonomous web scraping and interactive automation without local infrastructure.

Can I use Playwright and Puppeteer for autonomous web navigation with LLM assistance?▼

LLM-assisted automation uses vision-driven decision making to guide Playwright and Puppeteer execution, enabling dynamic web navigation, form filling, and interactive automation across cloud browser sessions.

Does browser automation work with CAPTCHA solving and live view in cloud sessions?▼

Cloud browser sessions support integrated CAPTCHA solving and live view monitoring, enabling autonomous web automation workflows to bypass challenges while providing real-time visual oversight.

What is the best way to manage browser session lifecycle and proxies for automated testing?▼

Browser session lifecycle and proxy management are handled natively through Steel, Browserbase, and Browser Use cloud providers, which configure environment keys, maintain session state, and route automation traffic.

Do I need Selenium to perform vision-assisted automation across cloud browsers?▼

Vision-assisted automation uses CDP-based Playwright connections rather than Selenium, leveraging LLM-driven visual decision making to orchestrate headful cloud browser interactions and data extraction.