Linkedin MCP Server README

LinkedInMCP: Revolutionizing LinkedIn API Interactions

🚀 Comprehensive Setup Guide

Prerequisites

• Node.js 16+
• LinkedIn Developer Account
• MCP-Compatible LLM (Claude, OpenAI, etc.)

⚠️ LinkedIn API access limitations (read this first)

This MCP server talks to LinkedIn's official HTTP APIs. Several things people run into are LinkedIn product / partner restrictions, not bugs in the install steps:

1. client_credentials (2-legged OAuth) is not available to most developer apps.
   LinkedIn docs state that apps cannot mint application tokens by default. A standard app that posts grant_type=client_credentials typically gets: access_denied / "This application is not allowed to create application tokens".
   That flow is reserved for approved LinkedIn enterprise / partner products. You usually need to contact a LinkedIn Relationship Manager or Business Development contact, or use the member authorization-code (3-legged) flow instead. See Client Credential Flow (2-legged OAuth).

2. There is no public general people-search endpoint at /search/people.
   LinkedIn does not document a free, unrestricted "search any member by keyword / company / location" API for ordinary developer apps. The search-people tool in this repo currently calls /v2/search/people, which is not a documented public endpoint and will not work with a standard developer token.
   What LinkedIn does document today is limited profile / people APIs (mostly for the authenticated member) and a restricted People Typeahead API for organization-page @mention search of page followers (docs). Broader people search requires a relevant LinkedIn partner product (for example Talent / Sales). Treat search-people as experimental / partner-only until this server is updated to a documented endpoint.

3. Send credentials in the POST body, never in the URL.
   Token requests now use Content-Type: application/x-www-form-urlencoded with client_id / client_secret in the body, matching LinkedIn's OAuth guidance, so secrets are not appended as query parameters.

1. LinkedIn Developer Setup

1. Go to LinkedIn Developer Portal
2. Create a new application
3. Obtain Client ID and Client Secret

Important: For detailed instructions on obtaining access tokens and authentication for LinkedIn API, refer to the official documentation: Getting Access to LinkedIn API

2. Project Installation

git clone https://github.com/Dishant27/linkedin-mcp-server.git
cd linkedin-mcp-server
npm install

3. Configuration Files

.env File

Create a .env file in the project root:

LINKEDIN_CLIENT_ID=your_client_id_here
LINKEDIN_CLIENT_SECRET=your_client_secret_here

claude_desktop_config.json (for Claude Desktop)

{
  "mcpServers": {
    "linkedin": {
      "command": "node",
      "args": ["/absolute/path/to/linkedin-mcp-server/dist/index.js"],
      "env": {
        "LINKEDIN_CLIENT_ID": "your_client_id",
        "LINKEDIN_CLIENT_SECRET": "your_client_secret"
      }
    }
  }
}

4. LLM Configuration Example

Claude.ai / Claude Desktop MCP Integration

{
  "tools": [
    {
      "name": "search-people",
      "description": "Search for LinkedIn profiles",
      "parameters": {
        "type": "object",
        "properties": {
          "keywords": {
            "type": "string",
            "description": "Keywords to search for in profiles"
          },
          "currentCompany": {
            "type": "array",
            "items": {"type": "string"},
            "description": "Filter by current company"
          },
          "industries": {
            "type": "array",
            "items": {"type": "string"},
            "description": "Filter by industries"
          },
          "location": {
            "type": "string", 
            "description": "Filter by location"
          }
        }
      }
    },
    {
      "name": "get-profile",
      "description": "Retrieve detailed LinkedIn profile information",
      "parameters": {
        "type": "object",
        "properties": {
          "publicId": {
            "type": "string",
            "description": "Public ID of the LinkedIn profile"
          },
          "urnId": {
            "type": "string", 
            "description": "URN ID of the LinkedIn profile"
          }
        }
      }
    }
  ]
}

5. Typical Workflow

1. Start the MCP Server:

npm run build
npm start

2. Example LLM Interactions:

> Find software engineers in San Francisco working at tech companies

# The LLM will use the search-people tool to:
# 1. Search LinkedIn profiles
# 2. Filter by location (San Francisco)
# 3. Filter by industry (Technology)
# 4. Return relevant profile details

📖 Learn More

Check out the detailed article explaining the concept and implementation:

LinkedIn MCP Server for LLMs

🚀 Project Background

LinkedInMCP is an innovative Model Context Protocol (MCP) server designed to transform how developers interact with LinkedIn's API. Born from the need for more flexible and powerful LinkedIn data integration, this project provides a robust, extensible framework for advanced LinkedIn data retrieval and interaction.

🌟 Project Vision

Recognizing the limitations of traditional LinkedIn API approaches, this solution creates a more intelligent, context-aware method of working with professional network data.

✨ Key Innovations

• Intelligent Context Management: Leverage MCP's advanced context handling
• Secure, Typed Interactions: Type-safe LinkedIn API interactions
• Extensible Architecture: Easy to expand and customize
• Standardized Protocol: Follows Model Context Protocol specifications

🛠️ Technologies

• TypeScript
• Model Context Protocol
• LinkedIn API
• OAuth 2.0
• Zod for Type Validation
• Axios for HTTP Requests

🔍 Core Capabilities

• Advanced People Search
• Detailed Profile Retrieval
• Job Market Intelligence
• Messaging Capabilities
• Secure Authentication Workflow

💼 Enterprise Use Cases

LinkedInMCP enables powerful applications for business settings:

Talent Acquisition

• Candidate Sourcing: Find qualified candidates matching specific skill and experience criteria
• Talent Pool Analysis: Map talent distribution across industries and geographies
• Outreach Automation: Streamline initial contact with potential candidates
• Competitive Intelligence: Understand recruitment patterns at competitor organizations

Sales & Business Development

• Lead Generation: Identify decision-makers at target companies
• Account-Based Marketing: Build comprehensive profiles of key accounts
• Relationship Mapping: Visualize connection networks within organizations
• Market Penetration Analysis: Track industry adoption of products or services

Market Research

• Industry Trends: Monitor employment shifts across sectors
• Skill Demand Analysis: Track emerging skills and technologies
• Organizational Structure Insights: Map reporting relationships and team compositions
• Geographic Movement Patterns: Understand talent migration between regions

Integration Capabilities

• CRM Synchronization: Update contact data from LinkedIn to CRM systems
• ATS Enhancement: Enrich applicant tracking systems with LinkedIn data
• Business Intelligence Tools: Feed LinkedIn data to dashboards and analytics platforms
• Custom Workflow Integration: Add LinkedIn capabilities to internal tools via API

🔒 Security Framework

The comprehensive security model of LinkedInMCP implements multiple layers of protection:

Authentication Security

• OAuth 2.0 Implementation: Industry-standard token-based authentication
• Token Refresh Management: Automated handling of token expiration
• Credential Isolation: Environment-based secret management
• Rate Limiting: Protection against excessive API requests

Data Security

• End-to-End Encryption: Secure data transmission
• Minimal Data Storage: Processing data without persistent storage
• Selective Information Return: Only returning requested data fields
• Automatic Data Sanitization: Preventing injection vulnerabilities

Compliance Features

• GDPR Adherence: Configurable data handling for European compliance
• API Terms Enforcement: Built-in LinkedIn API usage policy compliance
• Audit Logging: Comprehensive activity tracking for compliance reporting
• Data Governance Tools: Simplifying regulatory compliance management

📦 Generated Project Structure

linkedin-mcp-server/
│
├── src/
│   ├── index.ts         # Main server entry point
│   ├── auth.ts          # LinkedIn authentication handler
│   └── client.ts        # LinkedIn API interaction client
│
├── dist/                # Compiled JavaScript files
│   ├── index.js
│   ├── auth.js
│   └── client.js
│
├── .env                 # Secret environment variables
├── .env.example         # Template for environment variables
├── package.json         # Project dependencies and scripts
├── tsconfig.json        # TypeScript configuration
└── README.md            # Project documentation

🤝 Contributing

Inspired by the project? Contributions are welcome!

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Open a Pull Request

📄 License

Distributed under the MIT License.

Disclaimer: This project is an independent innovation and is not officially affiliated with LinkedIn or Microsoft.