PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteSome links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
To use Browser Use Cloud MCP, get an API key from your Browser Use account settings, then configure your MCP client to connect to https://api.browser-use.com/v3/mcp and send the key in the x-browser-use-api-key header. The exact configuration depends on your client: Claude Code, Claude Desktop, Cursor, and Windsurf have different documented formats. After setup, confirm that the server connects and its tools appear in the client.
What Browser Use MCP is—and which project this guide covers
This guide covers Browser Use Cloud MCP: a hosted browser-automation service with a remote MCP endpoint and Browser Use API-key authentication. Its endpoint, as shown in the Browser Use MCP guide, is https://api.browser-use.com/v3/mcp.
“Browser Use MCP” is easy to confuse with other browser-related MCP servers. Microsoft Playwright MCP and Browser MCP are separate projects with different setup methods and operating models. The instructions below are not interchangeable with theirs.
- Browser Use Cloud MCP: connect to Browser Use’s hosted endpoint and provide a Browser Use API key in a request header.
- Microsoft Playwright MCP: its official repository documents a locally launched setup using
npx @playwright/mcp@latest. Its getting-started guide describes browser interaction using structured accessibility snapshots, with browser-selection and profile options. - Browser MCP: its Copilot installation guide describes a Chrome extension and an MCP configuration using the
serverskey and@agent360/browser-mcp@latest.
If you intend to use Browser Use Cloud MCP, use its hosted URL and its header name—not a local launch command or another project’s configuration.
#1 Best Overall
Before you configure a client
Get an API key
The Browser Use guide directs users to obtain an API key through Browser Use account settings. Keep the key private: do not put a real key in a public repository, shared example, or other public configuration. The setup guide identifies the authentication header, but does not establish a complete key-management policy.
Identify your client’s expected configuration format
Browser Use supplies examples for Claude Code, Claude Desktop, Cursor, and Windsurf. Their syntax is not identical. In particular, a configuration that uses mcpServers and url is not a universal MCP configuration format; Windsurf’s Browser Use example uses serverUrl. Follow the current instructions for your particular client rather than pasting one client’s example into another.
Connect Browser Use MCP to an MCP client
Claude Desktop or Cursor: JSON example
This example follows the format shown for Claude Desktop and Cursor in the Browser Use guide. Add it using the configuration location and editing procedure documented by your client; the Browser Use page does not give one universal configuration-file path for every client.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →{
"mcpServers": {
"browser-use": {
"url": "https://api.browser-use.com/v3/mcp",
"headers": {
"x-browser-use-api-key": "YOUR_API_KEY"
}
}
}
}Replace YOUR_API_KEY with the key from your Browser Use account. Preserve the exact header name, x-browser-use-api-key, and the endpoint path, /v3/mcp. Do not commit a configuration containing a working key to a shared or public project.
Rank #2
- HTML CSS Design and Build Web Sites
- Comes with secure packaging
- It can be a gift option
Claude Code: use its HTTP MCP connection format
The Browser Use example for Claude Code uses an HTTP MCP connection and passes the API key as the x-browser-use-api-key header. Use the current Claude Code command or configuration syntax shown in the official Browser Use guide and Claude Code’s own documentation. Do not substitute the JSON example above unless your client’s documented schema calls for mcpServers, url, and headers.
Windsurf: use the documented serverUrl variant
Browser Use’s Windsurf example uses serverUrl and the same x-browser-use-api-key header. That field differs from the url field in its Claude Desktop and Cursor examples. Follow the Windsurf-specific example in Browser Use’s guide; changing the field name to match another client can prevent the server from connecting.
Check the connection
- Save the MCP configuration using the method required by your client.
- Use the client’s documented process to reload, restart, or reconnect its MCP servers, if needed.
- Check the client’s server or tool panel for a successful Browser Use connection and the Browser Use tools.
- If the server does not connect, check the endpoint, configuration schema, and API-key header before changing anything else.
The Browser Use documentation provides client examples, but it does not supply one configuration path or one reload procedure that applies to every client and version.
What the Browser Use tools do
The current Browser Use guide lists session-oriented tools for starting browser work, checking results, and managing sessions. Tool names and offerings can change, so use the names your connected client actually exposes.
Rank #3
| Tool | Documented purpose | When you might use it |
|---|---|---|
run_session |
Create a session and run a task. Listed options include keep_alive, model selection, an output schema, and browser profile ID. |
Start a new browser task and, where applicable, specify how its session should be configured. |
get_session |
Poll session status and output, including status, step count, cost breakdown, and live URL. | Check whether work is still running and inspect the returned session information. |
send_task |
Send a follow-up task to an idle keep-alive session. | Continue work in a session that remains available for another task. |
stop_session |
Stop a task or destroy the sandbox, depending on the selected strategy. | End work or tear down the session according to its configured behavior. |
get_session_messages |
Retrieve the agent’s messages, including browser actions, reasoning, and results. | Review the messages associated with a session. |
list_sessions |
List recent sessions with status and cost. | Review recent browser sessions. |
list_browser_profiles |
List browser profiles for authenticated tasks. | Find profiles available for tasks that require authentication. |
The presence of a cost breakdown in a tool description does not tell you the service’s price. The setup page does not state a current price schedule or quota.
Common setup problems and fixes
The MCP server does not connect
- Check the endpoint: use
https://api.browser-use.com/v3/mcp, including the/v3/mcppath. - Check the client schema: verify whether your client expects
url,serverUrl, or a different documented field. Use the format for that client. - Check JSON syntax: for a JSON-based configuration, look for missing commas, mismatched braces, or typographic quotes.
- Reload the client: apply changes using the client’s documented reload or restart process.
Authentication fails
- Confirm the key is present in a header named exactly
x-browser-use-api-key. - Check that you replaced the example text with your account’s actual API key and did not accidentally add spaces or quotes as part of the key value.
- If you changed or replaced the key in account settings, update the client configuration to match.
The connection works, but tools are missing
Confirm that the client has loaded the Browser Use server rather than another browser MCP project with a similar name. Then reload the server using the client’s own procedure and check its MCP tool list. The Browser Use guide’s list is a useful reference, but the available tools may change over time.
A task runs, but its result is unclear
Use the session-management tools shown by your client: get_session can poll status and output, while get_session_messages retrieves the session’s messages. The documented list_sessions tool can help locate recent sessions. The guide describes these purposes but does not promise that every client will present results in the same interface.
Cost, quotas, privacy, and operational expectations
The Browser Use MCP setup page documents the endpoint, API-key header, client examples, and tool purposes. It does not state current pricing, quotas, account eligibility, or a data-retention policy. Check Browser Use’s current account and service documentation for those terms rather than inferring them from a cost-related tool field.
Rank #4
- Brand: Wiley
- Set of 2 Volumes
- A handy two-book set that uniquely combines related technologies Highly visual format and accessible language makes these books highly effective learning tools Perfect for beginning web designers and front-end developers
The documented model is a hosted connection: your client connects to Browser Use’s endpoint. That is a concrete distinction from a locally launched server, but it does not by itself establish comparative privacy, speed, or reliability. The cited setup materials do not support a universal claim on those points.
For other MCP clients, configuration details depend on that client’s supported transport and schema. GitHub’s repository-level Copilot MCP configuration guide, for example, describes JSON configuration for its Copilot use case and warns that configured server tools can be used autonomously by Copilot in that environment. Treat that warning as specific to its described Copilot context, not as a universal behavior of every MCP client.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Or skip the browser setup
If your goal is to capture a clean webpage screenshot or PDF rather than automate an interactive browser session, ScreenshotNeo is a different tool: it is a website screenshot API and MCP server for developers, not a Browser Use MCP replacement for general browser tasks. A single GET request can return a PNG, JPEG, WebP, or PDF. See the ScreenshotNeo API documentation for the available parameters.
Recommended Free Tools
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpScreenshotNeo accepts cookie and consent banners like a visitor and removes more than 60 known consent platforms, newsletter popups, and chat widgets before capture; each step can be turned off. Bot checks, blank pages, timeouts, failed loads, and cache hits are not billed, and responses include X-Page-Verdict and X-Billed headers. Its MCP server gives AI agents tools named take_screenshot, get_page_info, and capture_pdf. The Free plan includes 1,000 screenshots per month with no card; paid plans start at $5 for 3,000 shots. Sign up for ScreenshotNeo’s free plan to get 1,000 screenshots a month with no card.
Which setup should you choose?
Use Browser Use Cloud MCP when you want the session-oriented browser tools documented by Browser Use and are prepared to connect your MCP client to its hosted endpoint with a Browser Use API key. Use a different project’s setup only when you specifically intend to use that project: Microsoft Playwright MCP’s local launch instructions and Browser MCP’s extension-backed setup are not Browser Use configuration alternatives.
If you only need a rendered page capture, use a screenshot tool instead of configuring a browser-automation session. A screenshot API is suited to producing an image or PDF; it is not a substitute for the broader session tasks listed for Browser Use.
Frequently Asked Questions
Does Browser Use MCP work with every MCP client?
The Browser Use guide provides examples for Claude Code, Claude Desktop, Cursor, and Windsurf. For other clients, compatibility and configuration syntax are not established by those examples; check the client’s current MCP documentation.
Can I use Microsoft Playwright MCP instructions to configure Browser Use?
No. Microsoft Playwright MCP is a separate project with its own setup and browser interaction model.
Does Browser Use MCP expose a tool for screenshots?
The Browser Use guide lists session-management tools, not a dedicated screenshot tool. For direct webpage captures, a screenshot API such as ScreenshotNeo is a separate option.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
