Hyperbrowser MCP Server

Name: Hyperbrowser MCP Server
Author: hyperbrowserai

@hyperbrowserai

Repository

About Hyperbrowser MCP Server

A MCP server implementation for hyperbrowser

#hyperbrowser #web-scraping #data-extraction

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

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

Tools

Extract formatted (markdown, screenshot etc) content from any webpage

Navigate through multiple linked pages and extract LLM-friendly formatted content

Convert messy HTML into structured JSON

Query the web and get results with Bing search

Fast, lightweight browser automation with the Browser Use agent

General-purpose automation using OpenAI’s CUA model

Complex browser tasks using Claude computer use

Creates a new persistent Hyperbrowser profile.

Deletes an existing persistent Hyperbrowser profile.

Lists existing persistent Hyperbrowser profiles.

Overview

What is Hyperbrowser MCP Server?

Hyperbrowser MCP Server is a Model Context Protocol server that provides tools to scrape, extract structured data, and crawl webpages. It also offers access to general‑purpose browser agents such as OpenAI's CUA, Anthropic's Claude Computer Use, and Browser Use.

How to use Hyperbrowser MCP Server?

Run npx hyperbrowser-mcp <YOUR-HYPERBROWSER-API-KEY>. For clients like Claude Desktop, Cursor, or Windsurf, add the configuration shown in the README to the corresponding MCP config file, passing the API key via the HYPERBROWSER_API_KEY environment variable.

Key features of Hyperbrowser MCP Server

Scrape webpages and extract formatted content (markdown, screenshot, etc.)
Crawl linked pages extracting LLM‑friendly content
Convert messy HTML into structured JSON
Query the web using Bing search
Run browser automation with Browser Use, OpenAI CUA, or Claude Computer Use agents
Create, delete, and list persistent browsing profiles

Use cases of Hyperbrowser MCP Server

Extract clean markdown or screenshots from any webpage
Crawl a site by following links and outputting structured content
Transform unstructured HTML into tidy JSON for data pipelines
Automate browser tasks with agent‑based computer use (CUA, Claude, or Browser Use)
Manage persistent browser profiles for repeated automation

FAQ from Hyperbrowser MCP Server

What do I need to run the server?

You need a Hyperbrowser API key and Node.js installed. The server is started via npx hyperbrowser-mcp <YOUR-API-KEY>.

How do I configure the server with Claude Desktop?

Add the example JSON from the README to your Claude Desktop MCP configuration file, replacing your-api-key with your actual API key.

What browser agents does the server support?

It supports OpenAI's CUA model, Anthropic's Claude Computer Use, and the Browser Use agent.

Which tools are available in the server?

The tools are: scrape_webpage, crawl_webpages, extract_structured_data, search_with_bing, browser_use_agent, openai_computer_use_agent, claude_computer_use_agent, create_profile, delete_profile, and list_profiles.