LinkedIn MCP Server

A Model Context Protocol (MCP) server that enables AI assistants to interact with LinkedIn's API through standardized tool calls. Compatible with Claude Code, Claude Desktop, and other MCP-enabled clients.

Features

• OAuth Authentication - Secure LinkedIn OAuth 2.0 flow
• Profile Management - Retrieve user profile information
• Post Creation - Create and publish LinkedIn posts with validation
• Content Guidelines - Built-in LinkedIn posting compliance
• Token Management - Automatic token storage and expiration handling
• Rate Limiting - Respects LinkedIn's API limits (1 post/minute, 100/day)

Quick Start

Prerequisites

• Python 3.8+
• uv package manager
• LinkedIn Developer App (see setup guide)

Installation

1. Clone the repository:

git clone <repository-url>
cd linkedin-mcp

2. Install dependencies:

npm run install
# or
uv sync

3. Configure environment variables:

cp .env.example .env
# Edit .env with your LinkedIn app credentials

Usage with Claude Code

Add the MCP server to Claude Code:

claude mcp add-json linkedin-mcp '{"type":"stdio","command":"uv","args":["--directory", "/path/to/linkedin-mcp", "run", "python", "main.py", "--stdio"]}'

Usage with Claude Desktop

Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "linkedin-mcp": {
      "command": "uv",
      "args": ["--directory", "/path/to/linkedin-mcp", "run", "python", "main.py", "--stdio"],
      "env": {
        "LINKEDIN_CLIENT_ID": "your_client_id",
        "LINKEDIN_CLIENT_SECRET": "your_client_secret",
        "LINKEDIN_REDIRECT_URI": "http://localhost:8000/linkedin/callback"
      }
    }
  }
}

Available Tools

Tool	Description
start_linkedin_auth	Generate OAuth authorization URL
exchange_linkedin_token	Exchange auth code for access token
check_linkedin_auth_status	Verify authentication status
get_linkedin_profile	Retrieve profile information
create_linkedin_post	Create and publish posts
linkedin_posting_guidelines	View content guidelines

Development

Running the Server

# OAuth callback server (for authentication)
python main.py

# MCP server (for Claude integration)
python main.py --stdio

# With uvicorn (development)
uvicorn main:app --reload --host 0.0.0.0 --port 8000

Code Quality

npm run lint      # Run ruff linting
npm run format    # Format code with ruff
npm run test      # Run pytest tests
npm run clean     # Clean Python cache files

Testing

# Local HTTP testing
curl http://localhost:8000/docs

# MCP testing
npm install -g @modelcontextprotocol/cli
mcp dev main.py --stdio

Architecture

• main.py - FastMCP server with unified MCP tools and OAuth callback
• linkedin_api.py - Direct LinkedIn API wrapper using requests
• database.py - SQLAlchemy-based token storage with async SQLite

LinkedIn API Compliance

This server enforces LinkedIn's content policies and technical limitations:

• Content Validation: 3000 character limit, 10 mentions maximum
• Rate Limiting: 1 post per minute, 100 posts per day
• Professional Standards: Content guidelines enforcement
• Token Management: Automatic expiration handling

Troubleshooting

Common Issues

1. OAuth Redirect Mismatch: Ensure redirect URI matches exactly in LinkedIn app settings
2. Permission Errors: Verify LinkedIn app has required products enabled
3. Token Expiration: Tokens expire after 60 days, re-authenticate when needed
4. Rate Limits: LinkedIn allows 1 post per minute, 100 posts per day

Getting Help

• Check the Tutorial for detailed setup instructions
• Review LinkedIn API documentation
• See MCP specification for protocol details

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Run tests and linting
5. Submit a pull request

License

MIT License

Acknowledgments

• Built with FastMCP
• LinkedIn API integration via python3-linkedin
• Model Context Protocol by Anthropic