Linked API

The Linked API MCP Server connects your AI assistant, like Claude or Cursor, directly to your LinkedIn account.

It works by giving the AI a set of tools to control LinkedIn through a secure cloud browser.

You can ask your assistant to perform tasks like searching for people, sending messages, or analyzing company profiles, and it will execute them for you.

Features

• 🔎 Advanced Search: Execute complex searches for people and companies with specific filters.
• 🤖 AI-Powered Actions: Let your AI send connection requests, messages, and post reactions on your behalf.
• 📊 Data Retrieval: Pull detailed information from personal profiles, company pages, and posts.
• 📈 Sales Navigator Support: Includes a separate set of tools for users with Sales Navigator accounts.
• 💬 Conversation Management: Your AI can read existing conversations and help you draft replies.
• ⚙️ Custom Workflows: Chain multiple actions together for complex automation sequences.

Use Cases

• Automated Lead Generation. You can give your AI a prompt like, “Find 10 software engineers at fintech companies in London with 50-200 employees.” The server will perform the search, and you can follow up by asking the AI to analyze their profiles and draft personalized connection notes for each one. This cuts down on hours of manual searching.
• Efficient Recruiting Outreach. Instead of manually searching for candidates, you can have your assistant find profiles with specific skills and experience. It can then handle the initial outreach, freeing you up to focus on conversations with interested candidates.
• Context-Aware Messaging. If you need to follow up with a list of contacts, you can have your AI read your past conversations with each person. It can then suggest a relevant follow-up message based on the existing context, which feels much more natural than a generic template.

How To Use It

1. The MCP server requires API tokens to link to your LinkedIn account.

• Sign up on the Linked API Platform.
• Select a plan and finish the purchase process.
• Follow the steps to connect your LinkedIn account.
• From your dashboard, copy the LINKED_API_TOKEN and the IDENTIFICATION_TOKEN.

If you manage more than one LinkedIn account, you will receive a unique identification token for each. This lets you run a separate MCP server instance for every account.

A quick note on the HEALTH_CHECK_PERIOD variable you’ll see in the configurations: some LinkedIn tasks take a few minutes. AI assistants often have short timeout limits. This setting sends a signal to keep the connection open during longer jobs, preventing the tool from disconnecting before your workflow is complete.

2. Configure your MCP clients:

Claude

• Mac: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "linkedapi": {
      "command": "npx",
      "args": ["-y", "linkedapi-mcp@latest"],
      "env": {
        "LINKED_API_TOKEN": "your-linked-api-token-here",
        "IDENTIFICATION_TOKEN": "your-identification-token-here",
        "HEALTH_CHECK_PERIOD": "180"
      }
    }
  }
}

Cursor

{
  "mcpServers": {
    "linkedapi": {
      "command": "npx",
      "args": ["-y", "linkedapi-mcp@latest"],
      "env": {
        "LINKED_API_TOKEN": "your-linked-api-token-here",
        "IDENTIFICATION_TOKEN": "your-identification-token-here",
        "HEALTH_CHECK_PERIOD": "600"
      }
    }
  }
}

You can also do this in the UI. Go to Cursor Settings > Tools & Integrations > MCP Tools Section, and click “New MCP Server” to paste the configuration.

VS Code

{
  "mcpServers": {
    "linkedapi": {
      "command": "npx",
      "args": ["-y", "linkedapi-mcp@latest"],
      "env": {
        "LINKED_API_TOKEN": "your-linked-api-token-here",
        "IDENTIFICATION_TOKEN": "your-identification-token-here",
        "HEALTH_CHECK_PERIOD": "600"
      }
    }
  }
}

Windsurf

Add the configuration to ~/.codeium/windsurf/mcp_config.json.

{
  "mcpServers": {
    "linkedapi": {
      "command": "npx",
      "args": ["-y", "linkedapi-mcp@latest"],
      "env": {
        "LINKED_API_TOKEN": "your-linked-api-token-here",
        "IDENTIFICATION_TOKEN": "your-identification-token-here",
        "HEALTH_CHECK_PERIOD": "600"
      }
    }
  }
}

This can also be configured through theUI in Windsurf Settings > Advanced Settings > Cascade section by clicking “Add Server”.

Available Tools

Standard

Sales Navigator

Other Tools

Usage Examples

Once configured, you can make requests in natural language:

• “Find all decision makers at Acme Corp and send them connection requests.”
• “Search for product managers at fintech companies in New York with 50-200 employees.”
• “Tell me about ‘https://linkedin.com/in/jane-doe’, including their work history and experience.”
• “Send a connection request to ‘https://linkedin.com/in/jane-doe’ and mention their recent article about AI in healthcare.”
• “Get all my pending connection requests and withdraw each one.”

Your AI can also combine these tools to execute more complex, multi-step workflows.

FAQs

Q: Is a paid Linked API account required?
A: Yes, the MCP server uses tokens that you get from a paid Linked API plan. These tokens authorize the server to act on your behalf.

Q: What is the purpose of the HEALTH_CHECK_PERIOD variable?
A: AI assistants often disconnect from a tool if it takes too long to respond. Some LinkedIn searches or workflows are slow. This setting keeps the connection active during these long-running tasks to prevent them from failing midway.

Q: Can I use this with more than one LinkedIn account?
A: Yes. Each LinkedIn account you connect to the Linked API platform gets its own unique IDENTIFICATION_TOKEN. You can run a separate MCP server configuration for each token.

Q: How does this interact with my LinkedIn account?
A: The service operates through a cloud browser to perform actions, which is a common method for automation. You should still be aware of LinkedIn’s usage policies to avoid sending too many requests in a short period.