playbooks

home / mcp / linkd mcp server

Linkd MCP Server

MCP implementation of https://linkd.inc/

Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "automcp-app-linkd-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "linkd-mcp"
      ],
      "env": {
        "LINKD_API_KEY": "YOUR-API-KEY"
      }
    }
  }
}

You can run the Linkd MCP Server to access LinkedIn data through a Model Context Protocol client. It lets you query users, companies, enrich profiles, retrieve contacts, scrape data, and perform deeper research tasks using a secure API key you provide.

How to use

Once the Linkd MCP Server is running, you connect your MCP client by configuring a local stdio server that launches the Linkd MCP runner with your API key. You can use different environments (Cursor, Windsurf, Claude Desktop) to start the server in the same way, with the API key supplied as an environment variable.

How to install

Prerequisites: Node.js and npm or npx must be available on your system.

Install and run the Linkd MCP Server using the following command. Replace YOUR-API-KEY with your actual API key.

npx linkd-mcp <YOUR-LINKD-API-KEY>

Additional setup for local MCP clients

You can config local MCP runners to start the Linkd MCP Server with the API key as an environment variable. The following configurations show how to wire the server into different MCP clients.

Cursor configuration example to run the server as a stdio process.

{
  "mcpServers": {
    "linkd": {
      "command": "npx",
      "args": ["-y", "linkd-mcp"],
      "env": {
        "LINKD_API_KEY": "YOUR-API-KEY"
      }
    }
  }
}

Usage with Windsurf and Claude Desktop

Windsurf and Claude Desktop can start the same MCP server by using equivalent configurations, including the API key in the environment.

{
  "mcpServers": {
    "linkd": {
      "command": "npx",
      "args": ["-y", "linkd-mcp"],
      "env": {
        "LINKD_API_KEY": "YOUR-API-KEY"
      }
    }
  }
}

Security notes

Keep your API key secret. Do not commit your keys to source control. Use environment variables or secret management for production deployments.

Troubleshooting

If you encounter connection issues, verify that the API key is set correctly in the environment and that the MCP client is configured to point at the same Linkd MCP Server instance.

Notes on available actions

The server exposes actions for common LinkedIn data tasks, including searching for users and companies, enriching LinkedIn profiles, retrieving contacts, scraping profiles and posts, and performing research at various depths.

Configuration

The Linkd MCP Server is configured to be started through a simple stdio-based runner. You supply your API key as an environment variable named LINKD_API_KEY. Each example above shows the exact shape you can reuse across different MCP clients.

License

The project is licensed under the MIT License.

Available tools

search_for_users

Search for LinkedIn users using filters such as query, school, and match threshold.

search_for_companies

Search for companies on Linkd using filters like query and match threshold.

enrich_linkedin

Retrieve detailed profile information for a specific LinkedIn URL (1 credit per lookup).

retrieve_contacts

Retrieve email addresses and phone numbers for a LinkedIn profile (1 credit per lookup).

scrape_linkedin

Retrieve detailed profile data and posts with comments from a LinkedIn profile URL (2 credits per request).

research_profile

Research a profile using email or phone number.

initiate_deep_research

Start a deep research job for comprehensive LinkedIn data gathering.

check_deep_research_status

Check the status of an ongoing deep research job.