🤖 Transform your AI applications with powerful web browsing capabilities! Let your AI read and understand the web.
Choose your favorite package manager:
# Using pip
pip install web-browser-mcp-server
# Using uv (recommended)
uv pip install web-browser-mcp-server
Add this to your Claude Desktop config to unlock web browsing superpowers:
{
"mcpServers": {
"web-browser-mcp-server": {
"command": "uv",
"args": [
"--directory",
"/path/to/web-browser-mcp-server",
"run",
"web-browser-mcp-server"
],
"env": {
"REQUEST_TIMEOUT": "30"
}
}
}
}
💡 Replace
/path/to/web-browser-mcp-server
with your installation path
Extract exactly what you need from any webpage:
# Basic webpage fetch
result = browse_webpage(url="https://example.com")
# Target specific content with CSS selectors
result = browse_webpage(
url="https://example.com",
selectors={
"headlines": "h1, h2",
"main_content": "article.content",
"navigation": "nav a"
}
)
Customize behavior with environment variables:
Variable | Description | Default |
---|---|---|
REQUEST_TIMEOUT | ⏱️ Max request time (seconds) | 30 |
USER_AGENT | 🕵️ Custom user agent string | Modern Chrome UA |
LOG_LEVEL | 📝 Logging verbosity | "info" |
MAX_RETRIES | 🔄 Max retry attempts | 3 |
Set up your dev environment in seconds:
# Create and activate virtual environment
uv venv
source .venv/bin/activate
# Install dev dependencies
uv pip install -e ".[test]"
# Run tests
python -m pytest
Contributions are welcome! Feel free to:
MIT License - do what you want! See LICENSE for details.
Built for the Model Context Protocol | Made with ❤️ by the MCP Community
Seamless access to top MCP servers powering the future of AI integration.