# Browserless

Connects AI agents to Browserless for browser control, web extraction, screenshots, PDFs, and Lighthouse checks. A Browserless API token is required.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 13
- Author: browserless
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/browserless/browserless-mcp
- Updated: 2026-10-23
- HTML: https://affpapa.org/en/nevibecode/mcp/browserless

## What it does

Browserless exposes a hosted browser through MCP. An agent can fetch one page, search the web, crawl a site, discover its URLs, and run Lighthouse audits. Page data can come back as Markdown or HTML, while the export tool handles PDFs, images, and other native formats.

The agent tool supports persistent sessions with clicks, typing, scrolling, tabs, uploads, downloads, and captcha handling. Saved authentication profiles can be reused across requests, and account tools show plan and unit details without exposing token values. You need a Browserless API token to connect; the HTTP setup does not require a local installation.

## Requirements
- Needs: Node.js (commands run via npx)
- Remote server: `https://mcp.browserless.io/mcp`
- Environment variables: `BROWSERLESS_TOKEN`, `BROWSERLESS_API_URL` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

1. **Platform CLI: remote server**
```bash
claude mcp add --transport http mcp https://mcp.browserless.io/mcp
```

2. **Platform CLI**
```bash
claude mcp add --env BROWSERLESS_TOKEN=<BROWSERLESS_TOKEN> --env BROWSERLESS_API_URL=<BROWSERLESS_API_URL> --transport stdio browserless -- npx -y @browserless.io/mcp
```
Replace <…> with your own values

### Codex

1. **Platform CLI: remote server**
```bash
codex mcp add browserless --url https://mcp.browserless.io/mcp
```

2. **Platform CLI**
```bash
codex mcp add browserless --env BROWSERLESS_TOKEN=<BROWSERLESS_TOKEN> --env BROWSERLESS_API_URL=<BROWSERLESS_API_URL> -- npx -y @browserless.io/mcp
```
Replace <…> with your own values

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "browserless": {
            "url": "https://mcp.browserless.io/mcp"
        }
    }
}
```
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "browserless": {
            "command": "npx",
            "args": [
                "-y",
                "@browserless.io/mcp"
            ],
            "env": {
                "BROWSERLESS_TOKEN": "<BROWSERLESS_TOKEN>",
                "BROWSERLESS_API_URL": "<BROWSERLESS_API_URL>"
            }
        }
    }
}
```
Replace <…> with your own values

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "browserless": {
            "type": "http",
            "url": "https://mcp.browserless.io/mcp"
        }
    }
}
```
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "browserless": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "@browserless.io/mcp"
            ],
            "env": {
                "BROWSERLESS_TOKEN": "<BROWSERLESS_TOKEN>",
                "BROWSERLESS_API_URL": "<BROWSERLESS_API_URL>"
            }
        }
    }
}
```
Replace <…> with your own values

### Gemini CLI

1. **Platform CLI: remote server**
```bash
gemini mcp add -t http browserless https://mcp.browserless.io/mcp
```

2. **Platform CLI**
```bash
gemini mcp add -e BROWSERLESS_TOKEN=<BROWSERLESS_TOKEN> -e BROWSERLESS_API_URL=<BROWSERLESS_API_URL> browserless npx -y @browserless.io/mcp
```
Replace <…> with your own values

## When to use it
- extract content from a JavaScript-heavy page
- crawl a site and collect its pages
- map a website’s URLs and descriptions
- run a Lighthouse audit for a URL
- export a webpage as a PDF or image

## Example prompts
- Read this page and return its content as Markdown.
- Crawl this site, keep pages under `/docs`, and return the scraped content.
- Run a Lighthouse check for this URL with a mobile device profile.

## Similar
- [browser](https://affpapa.org/en/nevibecode/skills/browser.md) — skill · 5900 installs
- [Browser Automation](https://affpapa.org/en/nevibecode/skills/claude-office-skills-browser-automation.md) — skill · 5200 installs
- [invisible_playwright_mcp](https://affpapa.org/en/nevibecode/mcp/invisible-playwright-mcp.md) — MCP server · ★ 884
- [Unbrowse](https://affpapa.org/en/nevibecode/mcp/unbrowse.md) — MCP server · ★ 810
- [browser-gateway](https://affpapa.org/en/nevibecode/mcp/browser-gateway.md) — MCP server · ★ 18
- [unbrowser by Unchained](https://affpapa.org/en/nevibecode/mcp/unbrowser-by-unchained.md) — MCP server · ★ 16

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md