linkedin-mcp-server: an MCP server that lets AI assistants access LinkedIn profiles, jobs, and messages via a local browser session

What it solves

This project provides a way for AI assistants (such as Claude) to interact with LinkedIn data. It bypasses the need for a formal API by allowing the AI to read and interact with LinkedIn through a user's own authenticated browser session, enabling the automation of profile lookups, job searches, and messaging.

How it works

It operates as a Model Context Protocol (MCP) server. It uses a headless browser (Patchright Chromium) to scrape and interact with the LinkedIn web interface. The server can either import existing session cookies from a locally installed browser (like Chrome or Brave) or prompt the user to log in manually via a browser window to establish a persistent session.

Who it’s for

It is designed for users of AI agents and MCP-compatible clients who want to integrate their professional network and job search activities into their AI workflow.

Highlights

• Comprehensive Toolset: Includes tools to get person and company profiles, search for jobs and people, read/send messages, and extract feed posts.
• Session Management: Supports importing sessions from multiple Chromium-based browsers to avoid manual login.
• Flexible Deployment: Can be run via uvx, as a Claude Desktop MCP Bundle, or within a Docker container.
• Crawl-based Access: Accesses data through the user's own logged-in session, providing a real-time view of LinkedIn as the user sees it.