Linkedin MCP Server

What is Linkedin MCP Server?

Linkedin MCP Server is an MCP server that enables interaction with the LinkedIn Community Management API. It uses HTTP+SSE transport for local or remote hosting and implements the draft Third-Party Authorization Flow from MCP specifications to delegate authorization to LinkedIn’s OAuth server.

How to use Linkedin MCP Server?

Install dependencies with pnpm install, copy .env.template to .env and populate with your LinkedIn client credentials and a JWT_SECRET. Run the server with pnpm run dev, then configure your MCP client to point to http://localhost:3001/sse.

Key features of Linkedin MCP Server

• user-info tool – get current user’s name, headline, and profile picture
• create-post tool – create a new post on LinkedIn
• Uses HTTP+SSE transport for local or remote deployment
• Implements draft Third-Party Authorization Flow via OAuth

Use cases of Linkedin MCP Server

• Automate LinkedIn post creation from an AI assistant
• Retrieve basic profile information of the authenticated user
• Integrate LinkedIn community management into an MCP-powered workflow

FAQ from Linkedin MCP Server

What does this server do?

It provides tools to interact with the LinkedIn Community Management API, specifically to get the current user’s info and create posts.

What are the runtime requirements?

Node 22 (lts/jod), pnpm 10, and a LinkedIn client with the Community Management API product installed, plus http://localhost:3001/callback as an authorized redirect URL.

How do I run the server locally?

After setting up the environment file, run pnpm run dev to start the server.

How can I debug the server?

Use the MCP Inspector by running pnpm run inspector and accessing http://localhost:5173 in a browser.

Is the authorization flow fully supported?

The Third-Party Authorization Flow is currently in draft. The only MCP client known to implement it is the MCP Inspector.