Hyperbrowser MCP

Browser & Desktop 💻 TypeScript ⚖️ MIT 🔴 No recent commits
789 stars

Hyperbrowser MCP is a Model Context Protocol server that provides tools to scrape, crawl, and extract structured data from webpages, as well as access to browser automation agents. It solves the problem of integrating web data extraction and browser automation into AI assistants and applications. It is for developers building with MCP-compatible clients like Claude Desktop, Cursor, or Windsurf.

✨ Key features

  • Scrape webpages to markdown or screenshots
  • Crawl multiple linked pages for content
  • Extract structured JSON from HTML
  • Search the web via Bing
  • Run browser automation agents (Browser Use, OpenAI CUA, Claude Computer Use)
  • Manage persistent browser profiles

🎯 Use cases

  • Extract product data from e-commerce sites
  • Monitor web content for changes
  • Automate form filling and data entry
  • Gather search results for research
  • Convert web pages into LLM-ready markdown

📦 Installation

🧰 Requirements: Requires Node.js and a Hyperbrowser API key.

Manual Installation

To install the server, run:

npx hyperbrowser-mcp <YOUR-HYPERBROWSER-API-KEY>

Running on Cursor

Add to ~/.cursor/mcp.json like this:

{
  "mcpServers": {
    "hyperbrowser": {
      "command": "npx",
      "args": ["-y", "hyperbrowser-mcp"],
      "env": {
        "HYPERBROWSER_API_KEY": "YOUR-API-KEY"
      }
    }
  }
}

Running on Windsurf

Add to your ./codeium/windsurf/model_config.json like this:

{
  "mcpServers": {
    "hyperbrowser": {
      "command": "npx",
      "args": ["-y", "hyperbrowser-mcp"],
      "env": {
        "HYPERBROWSER_API_KEY": "YOUR-API-KEY"
      }
    }
  }
}

Claude Desktop app

Example config:

{
  "mcpServers": {
    "hyperbrowser": {
      "command": "npx",
      "args": ["--yes", "hyperbrowser-mcp"],
      "env": {
        "HYPERBROWSER_API_KEY": "your-api-key"
      }
    }
  }
}

🚀 Usage

The server is typically configured as an MCP server in a client like Claude Desktop. After installation, you can use the tools provided by the server. For example, to scrape a webpage, you would invoke the scrape_webpage tool with the URL. The exact usage depends on the client. For development, you can run the server directly:

node dist/server.js

⚠️ Good to know

The MCP server provides a subset of the Hyperbrowser API features; the full API supports a superset.

❓ FAQ

What is Hyperbrowser MCP?

It is a Model Context Protocol server that provides tools for web scraping, crawling, and browser automation.

How do I install it?

You can run it with npx: npx hyperbrowser-mcp <YOUR-HYPERBROWSER-API-KEY> or configure it in MCP clients like Cursor, Windsurf, or Claude Desktop.

What tools are available?

Tools include scrape_webpage, crawl_webpages, extract_structured_data, search_with_bing, browser_use_agent, openai_computer_use_agent, claude_computer_use_agent, and profile management tools.

Do I need an API key?

Yes, you need a Hyperbrowser API key to use the server.

📊 Repository

Stars★ 789
Forks🍴 73
Open issues🐛 10
Last commit🕒 Nov 20, 2025
Created📅 Feb 2025
Language💻 TypeScript
License⚖️ MIT

🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.