platonai/Browser4

What is it?

Key features

• 01Drives real browsers through CDP, with headless and headed modes.
• 02Extracts data via X-SQL and CSS selectors from live pages or HTML snapshots.
• 03Converts saved HTML into spreadsheets and reports with WebMiner clustering, without LLM tokens.
• 04Runs swarm and crawl jobs with checkpoint-based resume.
• 05Exposes MCP over HTTP and supports plugins, runtime skills and a browser extension.

Use cases

• Letting an AI agent complete multi-step tasks such as comparing products on a shopping site.
• Extracting structured fields from many pages in a crawl without spending LLM tokens.
• Turning a folder of downloaded HTML pages into clustered spreadsheets with WebMiner.
• Automating repeated page checks on a schedule with loop jobs.

Quick start

1. 1

   Install browser4-cli globally with npm (requires Node.js)

   npm install -g browser4-cli
2. 2

   Install the Browser4 backend runtime

   browser4-cli install
3. 3

   Open a visible browser session to the example site

   browser4-cli open --headed https://browser4.io

Things to know

• !AI-powered commands such as extract, summarize, chat and agent run require an LLM provider API key.
• !The WebMiner clustering pipeline requires JDK 17 or newer.

In collections

Topics

FAQ

What is Browser4?

An AI-native browser engine for autonomous agents, intelligent data extraction and large-scale web automation, with a Rust CLI and MCP support. Browser4 is a browser engine built for AI agents and large-scale web automation. It drives real Chrome-based browsers through the Chrome DevTools Protocol, letting agents and humans navigate pages, click and fill elements, take snapshots, and run batch or looping jobs from a command-line tool called browser4-cli.

Is Browser4 free to use? What license does it have?

Browser4 is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.

Is Browser4 still maintained?

Actively maintained. The last commit to Browser4 was on Oct 22, 2026.

What can I use Browser4 for?

Letting an AI agent complete multi-step tasks such as comparing products on a shopping site. Extracting structured fields from many pages in a crawl without spending LLM tokens. Turning a folder of downloaded HTML pages into clustered spreadsheets with WebMiner. Automating repeated page checks on a schedule with loop jobs.

How do I install Browser4?

Install browser4-cli globally with npm (requires Node.js): npm install -g browser4-cli Install the Browser4 backend runtime: browser4-cli install Open a visible browser session to the example site: browser4-cli open --headed https://browser4.io

Where can I get Browser4?

The source code and installation instructions are on GitHub: https://github.com/platonai/Browser4