linkedin-mcp-server
This project is a Model Context Protocol server that provides interaction capabilities with LinkedIn, allowing users to scrape profiles, analyze companies, and perform job searches while securing user credentials.
What are the prerequisites for running the LinkedIn MCP Server?
You need Python 3.8 or higher, Chrome browser, ChromeDriver matching your Chrome version, and a LinkedIn account.
How are LinkedIn credentials stored?
Credentials are securely stored in your system's native keychain/credential manager, such as macOS Keychain or Windows Credential Locker.
What should I do if ChromeDriver errors occur?
Ensure your Chrome browser is updated, download the matching ChromeDriver version, set the CHROMEDRIVER path correctly, and try running with administrator/sudo privileges if needed.
How can I clear stored credentials?
Run the application with the --no-lazy-init flag and select 'Yes' to try with different credentials when prompted about login failure.
Is the LinkedIn MCP Server secure?
Yes, the server runs locally, and credentials are stored securely. Data scraping is done through your account, and profile visits are visible to other users.