Browser Ctl

A browser control MCP server based on Python Playwright and SSE Anthropic, providing a real-time web interaction toolset that supports functions such as JavaScript execution, URL access, and screenshot taking.

Browser automation Developer tools #Browser control #Web automation #Real-time interaction .Python

rating : 2 points

downloads : 12

update time : 2025-04-28

What is Browser CTL MCP Server?

Browser CTL is a Model Context Protocol (MCP) server that uses Playwright technology to enable AI agents to control and operate the Chrome browser like humans. Through simple API calls, AI can perform browser operations such as accessing web pages, executing JavaScript code, and taking screenshots.

How to use Browser CTL?

After installation, the server can be started through a simple command line. AI agents can communicate with the server through the SSE (Server-Sent Events) protocol and send instructions to control browser behavior.

Applicable scenarios

Suitable for AI application scenarios that require automated browser operations, such as web data scraping, automated testing, and web content monitoring. Particularly suitable for AI agents that require real-time browser interaction.

Main features

Execute JavaScriptExecute any JavaScript code in the browser in real-time and return the result

Web page navigationControl the browser to access a specified URL

Web page screenshotCapture the current browser page and return a base64-encoded image

Browser controlOpen/close the browser instance

Time controlSet the browser waiting time to simulate the rhythm of human operations

Advantages and limitations

Advantages

Based on Playwright technology, supporting all functions of modern browsers

Lightweight Python implementation, easy to deploy

Real-time communication through the SSE protocol

Providing complete browser automation capabilities

Limitations

Currently only supports the Chrome browser

Requires installation of Playwright dependencies

The SSE protocol may not be suitable for all application scenarios

How to use

Install dependencies

First, you need to install Python and project dependencies

Install the browser

Download the browser binary files required by Playwright

Start the server

Run the main program to start the MCP server

Connect and use

Connect to the server endpoint through the SSE protocol for interaction

Usage examples

Web page screenshotLet the AI agent get a screenshot of a specified web page

Data scrapingExtract specific data from a dynamic web page

Form auto-fillingAutomatically fill in and submit a web form

Frequently Asked Questions

How to modify the server port?

Which browsers are supported?

How to improve the operation speed?

Does it support multiple browser instances?

Related resources

Playwright official documentation

Official documentation for the Playwright Python library

SSE protocol introduction

Explanation of the Server-Sent Events protocol on MDN

BitEval YouTube channel

Video tutorials on how to create AI agents

GitHub repository

Project source code and latest updates

🚀 Browser CTL MCP Server

A browser control MCP server based on SSE and Anthropic, built with Python Playwright, providing a series of tools for AI agents to control and interact with web browsers. This simple version uses the Chrome browser.

🚀 Quick Start

This Browser CTL MCP Server offers a set of tools for AI agents to interact with web browsers. Below are the details of the exposed tools, default configurations, installation steps, and more.

✨ Features

Exposed Tools

Execute JavaScript: Execute JavaScript code in a live browser using Playwright's evaluation method and return the script result.
Access URL: Visit the target URL and open it in a live browser.
Wait with Delay: Have the browser wait for a specified random duration (between start and end times).
Close Browser: Close the opened browser window.
Open Browser: Launch the browser in a live environment.
Take Browser Screenshot: Capture a screenshot of the current browser page and return the image data in Base64 format.

Default Logging

This MCP server starts with the information logging level by default:

# Configure logging
logging.basicConfig(
    level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
)

Default Endpoint

http://127.0.0.1:8001/sse

📦 Installation

Step 1: Install uv and Set Up Python Project and Environment

Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Step 2: Clone the Repository

git clone https://github.com/biteval/browser_ctl.git

Step 3: Navigate to the Project Directory

cd browser_ctl

Step 4: Create and Activate a Virtual Environment

uv init
uv venv

Linux:

source .venv/bin/activate

Windows:

.venv\Scripts\activate

Step 5: Install Dependencies

Linux:

uv add "mcp[cli]" httpx playwright

Windows:

uv add mcp[cli] httpx playwright

Step 6: Download Playwright Browser Binaries

python -m playwright install

Step 7: Start the Browser CTL MCP Server

python browser_ctl.py

🤝 Contribution

Your expertise and creativity will have a significant impact on this project. Thank you for your participation!

ℹ️ More Information

Visit our YouTube channel to learn how to create AI agents:

Featured MCP Services

Duckduckgo MCP Server

Certified

The DuckDuckGo Search MCP Server provides web search and content scraping services for LLMs such as Claude.

The GitLab MCP server is a project based on the Model Context Protocol that provides a comprehensive toolset for interacting with GitLab accounts, including code review, merge request management, CI/CD configuration, and other functions.

TypeScript

4.3 points

Markdownify MCP

Markdownify is a multi-functional file conversion service that supports converting multiple formats such as PDFs, images, audio, and web page content into Markdown format.

A Python-based MCP Server that provides advanced to-do list management and content organization functions through the Notion API, enabling seamless integration between AI models and Notion.

UnityMCP is a Unity editor plugin that implements the Model Context Protocol (MCP), providing seamless integration between Unity and AI assistants, including real - time state monitoring, remote command execution, and log functions.

563

5 points

Figma Context MCP

Framelink Figma MCP Server is a server that provides access to Figma design data for AI programming tools (such as Cursor). By simplifying the Figma API response, it helps AI more accurately achieve one - click conversion from design to code.

A Gmail automatic authentication MCP server designed for Claude Desktop, supporting Gmail management through natural language interaction, including complete functions such as sending emails, label management, and batch operations.

Context7 MCP is a service that provides real-time, version-specific documentation and code examples for AI programming assistants. It is directly integrated into prompts through the Model Context Protocol to solve the problem of LLMs using outdated information.

TypeScript

5.2K

4.7 points

Zhiqi Future, Your AI Solution Think Tank

English 简体中文繁體中文にほんご