Skip to main content
BrowserClaw is a free, open-source browser your AI agents drive using your logged-in accounts. Your AI opens its own tabs inside it, and you watch, approve, and audit every step from a dashboard on your new-tab page.

​
Three things it gives your agent

A real browser

Your agent gets Chromium with your logins already there. No headless scraping, no fake user agents, no captchas that stop it dead.

A recording of every step

Every click, tab, and decision is captured. Scrub through the whole session frame by frame, or jump to the exact moment the agent chose wrong.

A cockpit that watches

A dashboard shows what your agent is doing right now and what it did before. No context-switching. No “did it finish?” pings.

​
Who this is for

Developers running Claude Code, Cursor, Codex, VS Code, Zed, OpenCode, or Antigravity in agentic mode. You want the agent to act on the web (log in somewhere, extract data, book a thing, test a flow) without giving it your own tabs and cookies. You want to see what it did after the fact.

​
Quick start

1

Install BrowserClaw

Download BrowserClaw for your operating system.

Install guide

macOS, Windows, and Linux downloads.
2

Connect your agent

Open the MCP install board inside BrowserClaw and click Install on your harness’s row. One click, no config file editing.

Connect your MCP client

Claude Code, Cursor, Codex, VS Code, Zed, OpenCode, Antigravity.
3

Paste the starter prompt

In your agent, paste Use BrowserClaw. Book me the cheapest morning flight from SFO to NYC next Friday. and press enter.

More sample prompts

A copyable library of prompts BrowserClaw handles well.
Wondering how BrowserClaw fits alongside BrowserOS? See BrowserClaw vs BrowserOS for the one-glance breakdown.