How to add Hyperbrowser MCP to Cursor

Hyperbrowser's cloud browser MCP — scrape, crawl, extract structured data, run CUA/Claude/Browser-Use agents. Paste the config into ~/.cursor/mcp.json and restart Cursor.

Last updated June 15, 2026 · 250★ · stdio · apikey · official

Cursor config for Hyperbrowser MCP

npx -y hyperbrowser-mcp

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

Setup steps

1Open Cursor → Settings → MCP → Add new MCP server (or edit ~/.cursor/mcp.json directly).
2Paste the Hyperbrowser MCP config below into the "mcpServers" object.
3Fill in placeholder secrets, then save.
4Cursor reloads MCP servers automatically — check Settings → MCP for a green status dot.
5Ask Cursor to use one of Hyperbrowser MCP's tools to confirm it's connected.

Before you start

Node.js installed (the server runs via npx)
A Hyperbrowser API key — sign up at https://app.hyperbrowser.ai and create a key in the dashboard (docs at https://docs.hyperbrowser.ai)
Set the key as the HYPERBROWSER_API_KEY environment variable (or pass it as a CLI argument)

What Hyperbrowser MCP can do in Cursor

scrape_webpage

Extract formatted content (markdown/HTML/links) from a single webpage.

crawl_webpages

Follow links across multiple pages of a site and extract their content.

extract_structured_data

Convert page HTML into structured JSON against a schema you provide.

search_with_bing

Run a web search via Bing and return ranked results.

browser_use_agent

Fast agentic browser automation powered by the Browser Use agent.

openai_computer_use_agent

Automate tasks using OpenAI's Computer-Use Agent (CUA) model.

claude_computer_use_agent

Run complex multi-step tasks via Claude computer use.

create_profile

Create a persistent Hyperbrowser profile to retain sessions/cookies.

Security

Requires a billable HYPERBROWSER_API_KEY; keep it in env and rotate if leaked. Scraping and autonomous browser agents execute against arbitrary sites in the cloud, so the extracted content should be treated as untrusted input to your model.

Hyperbrowser MCP + Cursor FAQ

Where is the Cursor config file?

Cursor reads MCP servers from ~/.cursor/mcp.json. Paste the Hyperbrowser MCP config there under the "mcpServers" key and restart the client.

Is Hyperbrowser MCP safe to use with Cursor?

How do I authenticate?

Set HYPERBROWSER_API_KEY in your MCP client config, or pass the key as a CLI argument to `npx hyperbrowser-mcp <KEY>`. Create the key in the Hyperbrowser dashboard.

Do I need to run a browser locally?

No. The browser runs in Hyperbrowser's cloud — the local npx process only relays MCP calls, so proxies, CAPTCHAs, and stealth are handled server-side.

Which clients does it work with?

Any MCP client over stdio. The README documents Cursor, Claude Desktop, and Windsurf; it can also be installed via Smithery.

View repo Full Hyperbrowser MCP page