@ofershap
基本情報
MCP server + Chrome extension that gives AI coding agents control of your real browser with existing sessions and logins. Works with Cursor, Claude, Windsurf, and any MCP client.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"real-browser": {
"command": "npx",
"args": [
"-y",
"real-browser-mcp"
]
}
}
}Navigate to a URL in the active browser tab
Click an element on the page using a ref from snapshot or a CSS selector
Type text into an input element
Scroll the page or an element. Supports pixel offsets, scrolling to elements, and named positions (top/bottom). Works with virtual scroll containers used by social media sites.
Press a keyboard key or combination (Enter, Escape, Tab, ArrowDown, etc). Supports modifiers like Ctrl+A, Cmd+C.
Wait for a condition: element to appear, element to disappear, or a fixed delay. Useful for SPAs and dynamic content.
Get an accessibility tree snapshot of the page. Returns element refs you can use with click, type, and other tools. Use compact mode (default) for smaller output - only interactive elements.
Capture a screenshot of the visible page area
Read console messages (log, warn, error) from the browser
Read network requests made by the page. Filter by URL pattern.
Manage browser tabs: list, create, close, or focus
Find elements on the page using natural language (e.g. "login button", "search input"). Returns refs you can use with click/type.
Extract raw text content from the page or a specific element
Hover over an element to trigger tooltips, dropdown menus, or hover states
Select an option from a dropdown/select element
Execute JavaScript in the page and return the result. Use for DOM queries, reading page state, or any operation not covered by other tools.
Click an element by its visible text content. Works on React dropdowns, portals, and overlays that may not appear in snapshots. CSP-safe (no eval). Prefers deepest matching element.
Handle JavaScript dialogs (alert, confirm, prompt). Dialogs block page interaction until handled.
Upload a file through a file input element. Uses CDP DOM.setFileInputFiles — works even on strict-CSP pages. Provide the local file path and either a ref from snapshot, a CSS selector targeting the file input, or omit both to auto-find the first input[type="file"].
Run a self-contained JavaScript action object in the page context via CDP. The code must be an expression that evaluates to an object with an execute(params) method. Returns the action result directly. Bypasses CSP restrictions.
Drag from one element or position to another. Uses CDP mouse events for reliable drag-and-drop. Provide either refs/selectors or explicit x,y coordinates.
Fill multiple form fields in a single call. Supports text inputs, selects, checkboxes, and contentEditable elements. Reduces round-trips compared to calling browser_type for each field individually. Optionally submit the form after filling.
Real Browser MCP is an MCP server paired with a Chrome extension that gives AI coding agents (Cursor, Claude, Windsurf) direct access to your actual browser — the one you already have open with all your sessions, logins, and cookies. Unlike tools that launch a fresh headless browser, it interacts with your existing browser environment.
Install the server globally via npx real-browser-mcp and install the companion Chrome extension from the Chrome Web Store. No further configuration is needed to start controlling your browser with AI agent prompts.
Unlike Playwright MCP or Chrome DevTools MCP, which launch a fresh headless browser, Real Browser MCP works with your actual browser. This means you never need to re-authenticate or replay login flows — your existing sessions, logins, and cookies are immediately available.
You need a Chrome browser and Node.js (to run `npx`). The server is installed via npm, and the extension is installed from the Chrome Web Store.
It works with Cursor, Claude, and Windsurf — any agent that implements the Model Context Protocol.
Run `npx real-browser-mcp` in your terminal and install the Chrome extension from [Chrome Web Store](https://chromewebstore.google.com/detail/real-browser-mcp/fkkimpklpgedomcheiojngaaaicmaidi).
🔮 Platform-neutral Browser Use for AI agents: real Chrome automation with a CLI + SDKs, no lock-in, dead simple.
Parse allows your agent to access thousands of APIs instantly. You can use Parse to build an API for any action on any website and edit any previously existing API in our catalogue.
Loadster is a load testing, stress testing, and synthetic monitoring platform. Run Playwright, Browser Bots, and Protocol Bots to test and monitor your websites, web apps, and APIs. Loadster MCP gives your agents the abi
An MCP server for generating images from HTML & CSS or screenshots of URLs using htmlcsstoimage.com.
コメント