@gtm-api/linkedin-mcp

Managed LinkedIn MCP server for AI agents. Stdio launcher for the hosted GTM API endpoint: search, connect, message and enrich on LinkedIn accounts you own. 160+ typed actions behind three MCP tools.

Source

npm

Version: 1.1.0

Version published: last month

Weekly downloads: 41

Maintainers: 1

Weekly downloads

Created: last month

Source

@gtm-api/linkedin-mcp

Launcher for GTM API: LinkedIn MCP Server, the managed LinkedIn MCP server for AI agents. It bridges a stdio MCP client (Claude Desktop, Cursor, anything that spawns a command) to the hosted streamable-http endpoint at https://mcp.gtm-api.com/mcp via mcp-remote. No server logic runs locally; every tool executes on the hosted server, where account safety is enforced: warm-up, server-side daily limits, preview-then-confirm on outward actions. GTM API reports 20,000+ LinkedIn accounts running at under 1% monthly ban; the method is written up at gtm-api.com/safe-linkedin-automation.

The server exposes three MCP tools (list_toolsets, get_toolset_tools, call_tool) that give an agent progressive access to 160+ typed LinkedIn actions: messaging, connection requests, content, enrichment, search, account health and infrastructure.

Setup

Get an API key at app.gtm-api.com (7-day trial, no card) and connect a LinkedIn account you own.
Add the server to your MCP client. Claude Desktop (claude_desktop_config.json):

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

Claude Code:

claude mcp add gtm-api -e GTM_API_KEY=YOUR_API_KEY -- npx -y @gtm-api/linkedin-mcp

Clients that speak streamable-http directly do not need this launcher; point them at https://mcp.gtm-api.com/mcp with an Authorization: Bearer YOUR_API_KEY header instead.