LinkedIn MCP Server

A Model Context Protocol (MCP) server for LinkedIn automation that provides three powerful tools for job searching, networking, and contact management.

🚀 Features

1. Job Search Tool (find_job_openings)

• Searches for the latest job openings based on your skills
• Filters by location, experience level, and job type
• Returns formatted job listings with company details and direct links

2. Auto Connect & Outreach Tool (auto_connect_and_outreach)

• Automatically finds and connects with relevant professionals
• Sends personalized outreach messages
• Targets specific roles and companies
• Rate-limited to avoid LinkedIn restrictions

3. Contact Scraper Tool (scrape_contacts_to_excel)

• Scrapes contact information from LinkedIn profiles
• Exports data to Excel spreadsheets
• Supports both direct profile URLs and search queries
• Collects names, titles, companies, and available contact info

📁 Project Structure

linkedin-mcp-server/
├── src/
│   ├── index.js                 # Main MCP server implementation
│   └── linkedin-automation.js   # Core LinkedIn automation logic
├── package.json                 # Dependencies and scripts
├── config.example.env          # Environment configuration template
└── README.md                   # This documentation

🛠️ Installation

Prerequisites

• Node.js 18.0.0 or higher
• npm or yarn
• A LinkedIn account

Setup Steps

1. Clone and install dependencies:

   npm install
   

2. Configure environment (optional):

   cp config.example.env .env
   # Edit .env with your preferences
   

3. Start the server:

   npm start
   

🔧 How It Works

Architecture Flow:

1. MCP Server Setup (src/index.js)

   • Initializes the MCP server with tool definitions
   • Handles incoming tool requests and routes them to appropriate functions
   • Returns formatted responses back to the MCP client

2. LinkedIn Automation (src/linkedin-automation.js)

   • Uses Puppeteer to control a Chrome browser
   • Handles LinkedIn login (manual for security)
   • Implements web scraping and interaction logic
   • Manages rate limiting and error handling

3. Tool Execution Flow:

   MCP Client → MCP Server → LinkedIn Automation → Browser → LinkedIn → Results
   

Security & Privacy:

• Manual login required - No password storage
• Rate limiting - Prevents account restrictions
• Headless mode option - For production use
• Error handling - Graceful failure management

📖 Tool Usage

1. Find Job Openings

{
  "name": "find_job_openings",
  "arguments": {
    "skills": ["JavaScript", "React", "Node.js"],
    "location": "San Francisco",
    "experience_level": "mid",
    "job_type": "full-time"
  }
}

Parameters:

• skills (required): Array of your skills to match against job postings
• location (optional): Preferred job location (default: "Remote")
• experience_level (optional): "entry", "mid", "senior", or "executive"
• job_type (optional): "full-time", "part-time", "contract", or "internship"

2. Auto Connect & Outreach

{
  "name": "auto_connect_and_outreach",
  "arguments": {
    "target_roles": ["Software Engineer", "Frontend Developer"],
    "company_names": ["Google", "Microsoft", "Apple"],
    "message_template": "Hi {name}, I'm interested in opportunities at {company}. Would love to connect!",
    "max_connections": 15
  }
}

Parameters:

• target_roles (required): Array of job titles/roles to target
• company_names (optional): Specific companies to focus on
• message_template (optional): Custom message with {name} and {company} placeholders
• max_connections (optional): Maximum connections to send (default: 10, max: 50)

3. Scrape Contacts to Excel

{
  "name": "scrape_contacts_to_excel",
  "arguments": {
    "search_query": "Product Manager at Tech Companies",
    "excel_filename": "product_managers.xlsx"
  }
}

Or with specific profile URLs:

{
  "name": "scrape_contacts_to_excel",
  "arguments": {
    "profile_urls": [
      "https://www.linkedin.com/in/johndoe/",
      "https://www.linkedin.com/in/janedoe/"
    ],
    "excel_filename": "specific_contacts.xlsx"
  }
}

Parameters:

• profile_urls (optional): Array of LinkedIn profile URLs to scrape
• excel_filename (optional): Name of Excel file to save (default: "linkedin_contacts.xlsx")
• search_query (optional): Search query to find profiles automatically

⚠️ Important Notes

LinkedIn Terms of Service

• This tool is for educational and personal use
• Respect LinkedIn's rate limits and terms of service
• Use responsibly to avoid account restrictions
• Consider LinkedIn's Premium features for enhanced functionality

Limitations

• Contact Information: Email/phone numbers are often protected by LinkedIn's privacy settings
• Rate Limiting: LinkedIn enforces strict limits on automated actions
• Selector Changes: LinkedIn may update their website structure, requiring code updates
• Account Security: Use manual login to maintain account security

Best Practices

1. Start Small: Test with low limits first (5-10 connections)
2. Add Delays: Respect rate limits to avoid detection
3. Personalize Messages: Use meaningful connection requests
4. Monitor Results: Check success rates and adjust accordingly
5. Stay Updated: LinkedIn may change their interface

🛠️ Development

Running in Development Mode

npm run dev

Testing

npm test

Browser Configuration

• Set HEADLESS_MODE=true in .env for production
• Browser will open for manual login on first run
• Session persists until server restart

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

📄 License

MIT License - See LICENSE file for details

🔮 Future Enhancements

• Support for LinkedIn Sales Navigator
• Advanced message templates with AI
• Integration with CRM systems
• Automated follow-up sequences
• Analytics and reporting dashboard
• Multi-account management
• Webhook integrations

Disclaimer: This tool is for educational purposes. Users are responsible for complying with LinkedIn's Terms of Service and applicable laws.