Skip to main content

MCP server for web browser automation using Julia Browser

Project description

julia-browser-mcp

Give any AI agent the ability to browse the web, click buttons, fill forms, and interact with websites—without Chrome or Firefox installation!

What is this?

This is an MCP (Model Context Protocol) server that lets AI assistants control a web browser. Think of it as giving your AI assistant hands to interact with websites the way you do.

Quick Start

Installation

Install from PyPI:

pip install julia-browser-mcp

For Claude Desktop Users

  1. Add to Claude Desktop:

    Open your Claude Desktop config file:

    • Mac: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json

    Add this configuration:

    {
      "mcpServers": {
        "webbrowser": {
          "command": "webbrowser-mcp"
        }
      }
    }
    
  2. Restart Claude Desktop

That's it! Claude can now browse websites for you.

What Can It Do?

Once connected, you can ask Claude things like:

  • "Open google.com and search for the latest news"
  • "Go to example.com and tell me what's on the page"
  • "Fill out this contact form with my information"
  • "Click the login button and enter my username"
  • "Scroll down and find the pricing section"

The AI assistant will use these tools behind the scenes:

Website Navigation

  • Open any website
  • Follow links
  • Get page content and titles

Interaction

  • Click buttons and links
  • Type into text fields
  • Submit forms

Page Analysis

  • List all clickable elements
  • Search for specific text
  • Get complete page information

Scrolling

  • Scroll up and down
  • Jump to top or bottom
  • Track scroll position

Using with Other MCP Clients

The server works with any MCP-compatible client. Just point it to the command:

webbrowser-mcp

The server uses stdio transport, so it's compatible with:

  • Claude Desktop
  • Cline (VS Code extension)
  • Any other MCP client that supports stdio

Example Conversation

You: "Go to example.com and tell me what you see"

Claude: Uses the open_website tool

"I've opened example.com. It's a simple example website with the heading 'Example Domain' and some text explaining that this domain is for use in examples."

You: "Are there any links I can click?"

Claude: Uses the list_elements tool

"Yes, there's a 'More information...' link on the page."

Requirements

  • Python 3.8 or higher
  • Node.js (required for the browser automation library)

These are automatically installed when you run pip install -e .

How It Works

The server uses the Julia Browser library to control a headless browser. When an AI assistant calls a tool (like "open website"), the browser performs the action and returns the results in a format the AI can understand.

All browser state is maintained across tool calls, so the AI can perform multi-step tasks like "go to this website, fill out the form, and submit it."

Troubleshooting

Claude doesn't see the tools:

  • Make sure you restarted Claude Desktop after editing the config
  • Check that the server is installed: run webbrowser-mcp in your terminal
  • Verify the config file path is correct for your operating system

Tools aren't working:

  • The browser needs internet access to visit websites
  • Some websites may block automated access
  • Try with simple websites first (like example.com) to verify it works

Safety Note

This tool allows AI assistants to browse the web and interact with websites. Always review what actions the AI is planning to take, especially on websites where you're logged in or entering sensitive information.

Author

Created by Harish Santhanalakshmi Ganesan

License

MIT License - feel free to use and modify as needed!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

julia_browser_mcp-1.0.0.tar.gz (5.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

julia_browser_mcp-1.0.0-py3-none-any.whl (5.9 kB view details)

Uploaded Python 3

File details

Details for the file julia_browser_mcp-1.0.0.tar.gz.

File metadata

  • Download URL: julia_browser_mcp-1.0.0.tar.gz
  • Upload date:
  • Size: 5.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.13

File hashes

Hashes for julia_browser_mcp-1.0.0.tar.gz
Algorithm Hash digest
SHA256 cfcdcb2e3e529105a617697bfaba6e6a5100329946c9b6b358c764c5cd6c8466
MD5 1cd4129a40c8d31e36c996c0f98cface
BLAKE2b-256 8d5850a4a939a60df2782ea0d3196009cff49b2ae70e453cc7e3d8f10f42f72b

See more details on using hashes here.

File details

Details for the file julia_browser_mcp-1.0.0-py3-none-any.whl.

File metadata

File hashes

Hashes for julia_browser_mcp-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 ffe7ee0ef72982c20c3f0bb69a7e4091ae8e6481a1deecb0a93e6868b4802024
MD5 5b3ae7671064c3281b8d45a0bdc8d7d0
BLAKE2b-256 a8d1ed0a3cbbf44afa52d19a3e66cb12a571b3a8f4a880232e488ae425cb1012

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page