BrowserAct Skills

BrowserAct Skills is an open-source browser automation Skill from the BrowserAct team. The README says it lets AI agents and teams automate a real browser, extract web data and run workflows that need an account. It connects to Chrome through a local command-line tool, either by importing your browser profile or by attaching to a Chrome that's already running, so it can reuse sites you're signed in to. When a step needs a person, it can send a live link to the page so someone can take over. The repo also includes Skill Forge and a set of ready-made solutions for specific websites. The docs are in English.

Who it's for:Developers who need an AI agent to operate a browser and know how to keep accounts and browsers separated. If you don't write code, you won't need it.

Before you start: The code is under the MIT licence. Plain Chrome mode needs no account; the "stealth" browsers and automatic CAPTCHA solving need a BrowserAct API key. The README says most of it is free, with charges beyond 5 stealth browsers and for hosted proxies; check the site for prices. The Skill documentation says cookies, credentials and page content stay on your computer, and that only the CAPTCHA image goes to an outside service, without naming it; the install docs also include a command that uploads logs. The Skill in the repo is only an entry point, and the full guide comes from running its command-line tool, which we didn't do. We don't recommend the features that spoof browser fingerprints, rotate proxies to get past anti-bot checks, solve CAPTCHAs automatically, reuse your signed-in sessions or act on your accounts for you.

Based on the project's public documentation. Not tested by AI Inspiration World.