Verified

Browserless MCP Server

Scrape JavaScript-heavy web content and perform searches via Browserless.io API.

Data last scanned today · Reviewed today

10 2 dangerous tools14 tools★ 7TypeScriptNOASSERTION

Overview

Browserless MCP Server bridges your AI assistant with the Browserless.io smart scraping infrastructure. By pointing your client to the hosted service using an API token, you gain the ability to process complex, dynamic websites that require JavaScript execution or anti-bot handling without managing local browser instances. The server provides tools for direct page scraping, web searching, site mapping, and multi-page crawling. It translates web content into formats suitable for LLM analysis, such as markdown or HTML, facilitating tasks like site auditing, content discovery, and targeted web research.

Our verdict

This server receives a high-risk rating with a security score of 40/100. The primary concerns stem from its status as a community-maintained project rather than an official vendor release, compounded by the use of an ambiguous NOASSERTION license and a lack of editorial review. Users should exercise caution as authentication methods and read-only tool implementations have not undergone security validation. It is suitable for testing purposes but should be monitored closely in any environment handling sensitive data.

Supports scraping JavaScript-heavy pages and anti-bot measures.
Small supply-chain surface with only 9 direct dependencies.
Features active development with a recent commit history.
Includes functional web mapping and multi-page crawling capabilities.

Carries a high-risk score of 40/100 due to unverified security.
Operates under an ambiguous NOASSERTION license.
Authentication and read-only protections are not yet reviewed.
Maintained by the community rather than the official Browserless organization.

Setup

Claude Desktop

{
  "mcpServers": {
    "browserless-mcp": {
      "args": [
        "-y",
        "@browserless.io/mcp"
      ],
      "command": "npx"
    }
  }
}

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart Claude Desktop. Some servers require API keys or tokens as environment variables — check the project's README and add an "env" object if needed.

No read-only mode — running this grants full read/write access.
No built-in authentication — anyone with access to this config can invoke every tool it exposes.

Tools

Tool	Description	Risk
browserless_smartscraper	Scrape a single webpage and return its content as markdown or HTML.	low
browserless_search	Search the web using Browserless and optionally scrape each result.	low
browserless_map	Discover and map all URLs on a website.	low
browserless_crawl	Crawl a website and scrape every discovered page.	low
browserless_performance	Run Lighthouse audits on any URL.	low
browserless_function	Execute custom Puppeteer JavaScript on the Browserless cloud.	high
browserless_export	Export a webpage via the Browserless /export API.	low
browserless_agent	Drive a persistent browser session via a ReAct loop.	high
browserless_skill	Load an on-demand recipe for a non-trivial page mechanic.	medium
browserless_profiles	List the authentication profiles saved for the current token.	low

Compatibility

Client	Local	Docker	Remote	Read-only
Claude Desktop
Cursor
VS Code
Windsurf
ChatGPT

Frequently asked questions

›Do I need to install software locally to use this server?

No, there is no local installation required. You configure the server by pointing your MCP client to the hosted URL provided in the configuration documentation.

›What is required to connect to the Browserless MCP server?

You need an API token obtained from browserless.io. A free tier is available for users.

›How do I reuse a logged-in state across different sessions?

You can use the browserless_profiles tool to list your saved authentication profiles and pass a profile name to other tools to reuse that logged-in state.

›Can I perform custom actions on the browser?

Yes, the browserless_function tool allows you to execute custom Puppeteer JavaScript on the Browserless cloud.

›How can I automate interactions with a website?

The browserless_agent tool drives a persistent browser session through a ReAct loop, supporting interactions like clicking, typing, scrolling, and multi-tab workflows.