# browser

This MCP server lets an AI agent use a real browser session to load authenticated pages, handle forms, and work with JavaScript-heavy sites.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 9
- Author: cherchyk
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/cherchyk/MCPBrowser
- Updated: 2026-10-12
- HTML: https://affpapa.org/en/nevibecode/mcp/browser

## What it does

\`browser` connects an AI agent to Chrome, Edge, or Brave through MCP. It can load pages, click elements, enter values into several form fields, read the current HTML, scroll, and capture screenshots. Because it uses your browser session, the agent can work with SSO, CAPTCHA checks, redirects, and pages that need JavaScript.

The server runs on Puppeteer while presenting the agent with structured results: cleaned HTML, SPA detection, and guidance for the next action. Tabs can be reused by domain, and the DOM can be read again without loading the page from scratch.

You need Chrome, Edge, or Brave plus Node.js 18+. Without Node.js 18+, the MCP server will not start.

## Requirements
- Needs: Node.js (commands run via npx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add browser -- npx -y mcpbrowser
```

### Codex

**Platform CLI**
```bash
codex mcp add browser -- npx -y mcpbrowser
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "browser": {
            "command": "npx",
            "args": [
                "-y",
                "mcpbrowser"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "browser": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "mcpbrowser"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add browser npx -y mcpbrowser
```

## When to use it
- open pages that require authentication
- handle SSO and redirect flows
- fill multi-step web forms
- read content from SPA pages
- inspect sites that require JavaScript

## Example prompts
- Open the sign-in page, complete authentication, and show me the contents of the account area.
- Find the form on the current page, fill it with values from `order.json`, and submit it.
- Browse the site, collect the requested page data, and save the result to `report.json`.

## Similar
- [Scrapling MCP Server](https://affpapa.org/en/nevibecode/mcp/scrapling-mcp-server.md) — MCP server · ★ 89899
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119231
- [AIHawk](https://affpapa.org/en/nevibecode/mcp/aihawk.md) — MCP server · ★ 32260
- [Skyvern](https://affpapa.org/en/nevibecode/mcp/skyvern.md) — MCP server · ★ 23283
- [SeleniumBase MCP](https://affpapa.org/en/nevibecode/mcp/seleniumbase-mcp.md) — MCP server · ★ 13181
- [apify-mcp-server](https://affpapa.org/en/nevibecode/mcp/apify-mcp-server.md) — MCP server · ★ 13345

---
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