zimejin07/linkedin-mcp-server
3.2
If you are the rightful owner of linkedin-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This server is designed to automate LinkedIn job searches while avoiding detection and maintaining session persistence.
Tools
1
Resources
0
Prompts
0
LinkedIn MCP Server
Description
A server to automate LinkedIn job searches with anti-detection features and persistent sessions.
Key Features
- ✅ Anti-detection: Uses puppeteer-extra with stealth plugin to avoid bot detection.
- ✅ Session persistence: Saves cookies so you only need to log in once.
- ✅ Headless mode: Runs invisibly in the background.
- ✅ Human-like behavior: Random delays, realistic typing speed.
- ✅ Security checkpoints: Handles LinkedIn's verification challenges.
- ✅ Persistent browser profile: Maintains login state across runs.
- ✅ Job search with filters: Filters by time posted, experience level, remote options.
- ✅ Detailed job info: Fetches full job descriptions.
Installation
# 1. Create project directory
git clone the repo
cd linkedin-mcp-server
# 2. cd linkedin-mcp-server
# 3. Install dependencies
npm install
# 4. Make it executable
chmod +x index.js
# 5. Set environment variables (secure way)
See below
Claude Desktop Configuration
Edit your config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
- Windows:
%APPDATA%\Claude\claude_desktop_config.json
Add this:
{
"mcpServers": {
"linkedin": {
"command": "node",
"args": ["/full/path/to/linkedin-mcp-server/index.js"],
"env": {
"LINKEDIN_EMAIL": "your-email@example.com",
"LINKEDIN_PASSWORD": "your-password"
}
}
}
}
Restart Claude Desktop.
Usage Examples
Once configured, you can prompt Claude:
- "Log into LinkedIn and search for Python developer jobs in New York posted in the last 24 hours."
- "Find remote software engineering jobs at the mid-senior level."
- "Get the details of this job: [LinkedIn job URL]."
Claude will automatically use the MCP tools to execute your requests.
Concerns Handled
- Bot Detection: Stealth plugin + realistic user agent.
- Rate Limiting: Random delays between actions.
- Session Management: Saves cookies, maintains browser profile.
- Security Checkpoints: Detects and alerts for manual verification.
- Credential Security: Uses environment variables.
- Headless Mode: Runs without a visible browser.
- Error Handling: Graceful failures with informative messages.
- Resource Cleanup: Proper browser closure.
First Time Setup
The first time you run this, LinkedIn may require verification (email/SMS). If this happens:
- The server will notify you.
- Set
headless: false
in the code temporarily. - Complete verification manually.
- Session will be saved for future automated use.