LinkedIn MCP config generator

1. In Claude Desktop open Settings, then Developer, and click Edit Config. That opens claude_desktop_config.json.
2. Paste the block. If the file already has an mcpServers object, add the gtm-api entry inside it rather than a second object.
3. Save and restart Claude Desktop. The launcher runs through npx, so the machine needs Node.js 18 or newer.

Where it goes

macOS

~/Library/Application Support/Claude/claude_desktop_config.json

Windows

%APPDATA%\Claude\claude_desktop_config.json

Copy this

{
  "mcpServers": {
    "gtm-api": {
      "command": "npx",
      "args": ["-y", "@gtm-api/linkedin-mcp"],
      "env": { "GTM_API_KEY": "gtm_live_..." }
    }
  }
}

1. Open Settings, then Connectors, and click Add custom connector.
2. Paste the server URL below and add it.
3. Approve the consent screen that opens in your browser. Nothing else to paste, and the same connector works on claude.ai and in Claude Desktop.

Copy this

https://mcp.gtm-api.com/mcp

1. Run the command in a terminal. With the key attached, the browser sign-in never starts.

Copy this

claude mcp add --transport http gtm-api https://mcp.gtm-api.com/mcp --header "Authorization: Bearer gtm_live_..."

Check from a terminal

claude mcp list

Shows gtm-api as connected.

1. Run the command in a terminal.
2. Start a Claude Code session, run /mcp and authenticate when prompted.

Copy this

claude mcp add --transport http gtm-api https://mcp.gtm-api.com/mcp

Check from a terminal

claude mcp list

Shows gtm-api as connected once the consent is done.

claude.ai connectors have no field for a header, so they always sign in through the browser. Your key is not used here.

Open connector settings

1. Open Settings, then Connectors, and click Add custom connector.
2. Paste the server URL below and add it.
3. Approve the consent screen that opens in your browser. Nothing else to paste, and the same connector works on claude.ai and in Claude Desktop.

Copy this

https://mcp.gtm-api.com/mcp

ChatGPT connectors sign in through the browser, so your key is not used here.

1. In ChatGPT open Plugins, select the plus button, then Add custom MCP server. Developer mode does not need to be on.
2. Give it a name, paste the server URL below under Connection and choose OAuth as the authentication.
3. Tick the risk acknowledgement, select Create as a plugin and complete the sign-in on the consent screen. On Business and Enterprise workspaces an admin may have to allow custom MCP servers first.

Copy this

https://mcp.gtm-api.com/mcp

1. Open mcp.json and paste the block. If the file already has an mcpServers object, add the gtm-api entry inside it.
2. Cursor starts the launcher through npx, so the machine needs Node.js 18 or newer.

Where it goes

All projects

~/.cursor/mcp.json

One project

.cursor/mcp.json

Copy this

{
  "mcpServers": {
    "gtm-api": {
      "command": "npx",
      "args": ["-y", "@gtm-api/linkedin-mcp"],
      "env": { "GTM_API_KEY": "gtm_live_..." }
    }
  }
}

Add to Cursor

1. Press Add to Cursor. It opens Cursor and adds the server.
2. Sign in on the consent page that follows.
3. Rather edit the file? The block is the same entry the button adds.

Where it goes

All projects

~/.cursor/mcp.json

One project

.cursor/mcp.json

Copy this

{
  "mcpServers": {
    "gtm-api": {
      "url": "https://mcp.gtm-api.com/mcp"
    }
  }
}

Our docs connect VS Code through the one-click install, which signs in through the browser. Your key is not used here.

Install in VS Code

1. Press Install in VS Code. It opens VS Code and adds the server.
2. Sign in on the consent page that follows.

Copy this

https://mcp.gtm-api.com/mcp

Our docs connect Codex with codex mcp login, a browser sign-in. Your key is not used here.

1. The first line registers the server in ~/.codex/config.toml.
2. The second runs the sign-in and opens the consent page. No client id to pass.

Copy this

codex mcp add gtm-api --url https://mcp.gtm-api.com/mcp
codex mcp login gtm-api

1. A client that takes a URL: give it the server URL over the streamable HTTP transport, and the header below. No browser, no consent screen.
2. A client that can only start a command: give it the launcher, which reads the key from GTM_API_KEY.

Copy this

https://mcp.gtm-api.com/mcp

Authorization: Bearer gtm_live_...

{
  "mcpServers": {
    "gtm-api": {
      "command": "npx",
      "args": ["-y", "@gtm-api/linkedin-mcp"],
      "env": { "GTM_API_KEY": "gtm_live_..." }
    }
  }
}

1. A client that takes a URL: give it the server URL over the streamable HTTP transport. It discovers the OAuth flow on its own, and dynamic client registration means nothing has to be registered with us first.
2. A client that can only start a command: give it the launcher. It opens the consent page on first use and caches the tokens under ~/.mcp-auth.

Copy this

https://mcp.gtm-api.com/mcp

{
  "mcpServers": {
    "gtm-api": {
      "command": "npx",
      "args": ["-y", "@gtm-api/linkedin-mcp"]
    }
  }
}

Check from a terminal

npx @modelcontextprotocol/inspector

Choose the streamable HTTP transport, enter the server URL, connect and approve the consent screen. The tool list appears once you are signed in.