LinkedIn MCP Server

Complete LinkedIn automation toolkit. Scrape profiles, manage posts, read any LinkedIn content, and automate interactions via MCP (Model Context Protocol).

---

📘 Table of Contents

• Overview
• Features
• Quickstart
• Authentication
• Usage
• Claude Desktop Integration
• Available Tools
• Project Structure
• Testing
• Troubleshooting
• API Versioning
• Contributing
• License
• Disclaimer
• Changelog

---

Overview

linkedin-mcp is a fully featured MCP server that provides automation tools for LinkedIn.
It supports browser-based scraping and API-based operations for content management, media uploads, and reactions.

Repository: Linkedin MCP Server

---

Features

🔍 Scraping (Browser-Based)

• Extract full LinkedIn profiles
• Scrape company pages
• Read job listings
• Read ANY LinkedIn post
• Extract images, videos, engagement metrics

📝 API-Based Post Management

• Create, update, delete LinkedIn posts
• Add or remove reactions
• Upload images and documents
• Supports all official LinkedIn REST API features

🧩 MCP Integration

• Works with Claude Desktop and any MCP-compatible client
• 17 total tools included

🧪 Testing

• 50+ tests
• Covers scraping, API, and MCP tools

---

Quickstart

Install

git clone https://github.com/selvin-paul-raj/Linkedin-MCP-Server.git
cd Linkedin-MCP-Server

# create environment config
cp .env.example .env

# install dependencies
pip install -e .

Run

# Standard MCP server
uv run linkedin-mcp

# Debug mode (shows browser)
uv run main.py --debug --no-headless --no-lazy-init

# HTTP mode
uv run main.py --transport streamable-http

---

Authentication

1. Scraping (LinkedIn Cookie)

Get your li_at cookie:

1. Log in to LinkedIn in Chrome
2. Press F12
3. Application → Cookies → https://www.linkedin.com
4. Copy the li_at cookie value
5. Add to .env:

LINKEDIN_COOKIE=li_at=YOUR_COOKIE_VALUE

---

2. API (OAuth Access Token)

Add these fields to .env:

LINKEDIN_CLIENT_ID=your_id
LINKEDIN_CLIENT_SECRET=your_secret
LINKEDIN_ACCESS_TOKEN=your_access_token
LINKEDIN_API_VERSION=202510

Quick OAuth Link

https://www.linkedin.com/oauth/v2/authorization?response_type=code&client_id=YOUR_CLIENT_ID&redirect_uri=YOUR_REDIRECT_URI&scope=w_member_social%20r_liteprofile%20r_emailaddress

Exchange auth code:

curl -X POST https://www.linkedin.com/oauth/v2/accessToken \
  -H "Content-Type: application/x-www-form-urlencoded" \
  -d "grant_type=authorization_code" \
  -d "code=YOUR_CODE" \
  -d "redirect_uri=YOUR_REDIRECT_URI" \
  -d "client_id=YOUR_CLIENT_ID" \
  -d "client_secret=YOUR_CLIENT_SECRET"

---

Usage

Read any LinkedIn post

{
  "tool": "read_linkedin_post",
  "input": "https://www.linkedin.com/posts/...activity-123456..."
}

Create a post

{
  "tool": "create_linkedin_post",
  "input": {
    "text": "Excited to announce our new product launch! 🚀",
    "visibility": "PUBLIC"
  }
}

Upload image

{
  "tool": "upload_linkedin_image",
  "input": { "image_url": "https://example.com/image.jpg" }
}

---

Claude Desktop Integration

Add this to claude_desktop_config.json:

{
  "mcpServers": {
    "linkedin": {
      "command": "uv",
      "args": [
        "--directory",
        "D:\\MCP\\linkedin-mcp",
        "run",
        "linkedin-mcp"
      ],
      "env": {
        "LINKEDIN_COOKIE": "li_at=YOUR_COOKIE",
        "LINKEDIN_ACCESS_TOKEN": "YOUR_TOKEN"
      }
    }
  }
}

---

Available Tools

📖 Content Reading

• read_linkedin_post

🌐 Scraping

• get_person_profile
• get_company_profile
• get_job_details
• search_jobs
• search_recommended_jobs
• close_session

📝 Post Management

• create_linkedin_post
• update_linkedin_post
• delete_linkedin_post

🖼️ Media

• upload_linkedin_image
• get_linkedin_image

💙 Reactions

• add_linkedin_reaction
• remove_linkedin_reaction
• get_linkedin_reactions

👤 Profile & Auth

• get_linkedin_profile
• validate_linkedin_credentials

More details. See TOOLS_REFERENCE.md.

---

Project Structure

linkedin-mcp/
├── linkedin_mcp_server/
│   ├── server.py
│   ├── cli.py
│   ├── config/
│   ├── drivers/
│   └── tools/
├── tests/
│   ├── unit/
│   ├── integration/
├── scripts/
├── .env.example
├── pyproject.toml
├── README.md
└── TOOLS_REFERENCE.md

---

Testing

# unit tests
uv run pytest tests/unit -v

# integration tests
uv run pytest tests/integration -v

# all tests
uv run pytest tests/ -v

Quick API test:

uv run python scripts/test_api.py

---

Troubleshooting

❌ "426 Client Error: Upgrade Required"

Fix:

LINKEDIN_API_VERSION=202510

❌ "LINKEDIN_COOKIE required"

Get fresh cookie from Chrome.

❌ "401 Unauthorized"

Generate a new access token.

ChromeDriver issues

pip install --upgrade selenium webdriver-manager

---

API Versioning

Current default:

202510

Check latest: https://learn.microsoft.com/en-us/linkedin/marketing/versioning

Update:

LINKEDIN_API_VERSION=202511

Restart the server.

---

Contributing

git clone https://github.com/selvin-paul-raj/Linkedin-MCP-Server.git
cd Linkedin-MCP-Server
uv sync
uv run pytest tests/ -v
uv run ruff format .
uv run pre-commit run --all-files

Pull requests welcome.

---

License

MIT License. See the LICENSE file.

---

Disclaimer

This tool is for educational and automation purposes. Follow LinkedIn TOS, API terms, and usage limits. Use responsibly.

---

Built with ❤️ for LinkedIn automation