Browser-use-claude-mcp
About
A browser automation MCP server for AI models like Claude and Gemini 2.5, enabling web browsing capabilities through natural language
Details
- License
- MIT license
Explore
- Full browser automation (navigation, form filling, clicking, etc.)
- Web search capabilities
- Screenshot capture for visual understanding
- Content extraction and analysis
- Support for multiple AI providers:
- Google Gemini 2.5 (primary focus)
- Anthropic Claude
- OpenAI
- Image analysis (vision) capabilities
- AI-powered content analysis
- Written in TypeScript for maximum reliability
- Modular architecture with clean separation of concerns
- Comprehensive logging and error handling
- Easy configuration through environment variables
See INSTALL.md for detailed installation and setup instructions.
CHROME_PATH=
CHROME_USER_DATA=
CHROME_DEBUGGING_PORT=9222
browse_webpage
Navigate to a URL and extract its content
search_web
Perform a web search and return results
take_screenshot
Capture a screenshot of the current page
click_element
Click on an element by text or selector
fill_form
Fill out form fields with provided values
extract_content
Extract specific content from a webpage
analyze_content
AI-powered analysis of webpage content
| Tool Name | Description |
|-----------|-------------|
| browse_webpage | Navigate to a URL and extract its content |
| search_web | Perform a web search and return results |
| take_screenshot | Capture a screenshot of the current page |
| click_element | Click on an element by text or selector |
| fill_form | Fill out form fields with provided values |
| extract_content | Extract specific content from a webpage |
| analyze_content | AI-powered analysis of webpage content |
A browser automation MCP server for AI models like Claude and Gemini 2.5, enabling web browsing capabilities through natural language.
Overview
This project implements a Model Context Protocol (MCP) server that provides browser automation capabilities to AI models. It allows AI assistants to browse the web, interact with websites, and extract information using natural language commands.
Key Features
🌐 Browser Automation Features
- Full browser automation (navigation, form filling, clicking, etc.) - Web search capabilities - Screenshot capture for visual understanding - Content extraction and analysis🤖 AI Features
- Support for multiple AI providers: - Google Gemini 2.5 (primary focus) - Anthropic Claude - OpenAI - Image analysis (vision) capabilities - AI-powered content analysis🔧 Technical Features
- Written in TypeScript for maximum reliability - Modular architecture with clean separation of concerns - Comprehensive logging and error handling - Easy configuration through environment variablesAvailable Tools
| Tool Name | Description |
|-----------|-------------|
| browse_webpage | Navigate to a URL and extract its content |
| search_web | Perform a web search and return results |
| take_screenshot | Capture a screenshot of the current page |
| click_element | Click on an element by text or selector |
| fill_form | Fill out form fields with provided values |
| extract_content | Extract specific content from a webpage |
| analyze_content | AI-powered analysis of webpage content |
Getting Started
See INSTALL.md for detailed installation and setup instructions.
Quick Start
1. Clone the repository
git clone https://github.com/jasondsmith72/Browser-use-claude-mcp.git
cd Browser-use-claude-mcp
2. Install dependencies
npm install
3. Create a .env file (use .env.example as a template)
cp .env.example .env
4. Build the project
npm run build
5. Start the server
npm start
Configuration
The server can be configured through environment variables in your .env file:
```
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.



