n8n-nodes-gtm-api

n8n community nodes for gtm-api.com, the LinkedIn API and MCP server for AI agents. Run LinkedIn outreach on accounts you own from n8n workflows: send connection requests and messages, enrich people, run searches, and trigger on replies. Every action executes on the hosted platform, where account safety is enforced: warm-up ramps for fresh accounts, server-side daily limits, health-aware pacing. 20,000+ LinkedIn accounts run on that stack, and under 1% of them have ever been restricted. The method is written up at gtm-api.com/safe-linkedin-automation.

Nodes

• gtm-api: actions on your sender accounts.
  • Connection Request: Send (note support, optional no-note fallback, resend cooldowns respected server-side)
  • Message: Send (to a member by ln_id / sn_id, or into an existing conversation)
  • Person: Enrich (lite profile: full name, profile slug, LinkedIn IDs, connection degree)
  • People Search: Search by URL (paste a people search URL, get structured rows)
• gtm-api Trigger: fires on webhook events.
  • Message Received, Connection Request Accepted, Invitation Received
  • Registers the webhook on activation and removes it on deactivation. The n8n instance must be reachable over public https, because the platform refuses non-public target URLs.

Install

It is a verified community node: search for gtm-api in the nodes panel and install it from there, on n8n Cloud and self-hosted alike. A self-hosted instance can also install it by name: Settings, Community Nodes, Install, enter n8n-nodes-gtm-api.

Credentials

One field: the API key. Create it at app.gtm-api.com (forever free plan, no card) and connect a LinkedIn account you own.

Example workflow

Search a people-search URL, then invite every row from one sender. Paste into a blank canvas (Ctrl+V), then pick your credential and pick the sender in the Sender dropdown on the invite node.

{
  "nodes": [
    {
      "parameters": {
        "resource": "peopleSearch",
        "operation": "searchByUrl",
        "url": "https://www.linkedin.com/search/results/people/?keywords=head%20of%20growth"
      },
      "type": "n8n-nodes-gtm-api.gtmApi",
      "typeVersion": 1,
      "position": [0, 0],
      "name": "Search people"
    },
    {
      "parameters": {
        "resource": "connectionRequest",
        "operation": "send",
        "profileId": "={{ $json.ln_id }}",
        "note": "=Hi {{ $json.full_name.split(' ')[0] }}, saw your work on outbound. Worth a chat?",
        "allowNoNoteFallback": true
      },
      "type": "n8n-nodes-gtm-api.gtmApi",
      "typeVersion": 1,
      "position": [240, 0],
      "name": "Send invite"
    }
  ],
  "connections": {
    "Search people": { "main": [[{ "node": "Send invite", "type": "main", "index": 0 }]] }
  }
}

No wait node is needed between the two. Pacing against that account's own daily budget, the warm-up ramp and the resend cooldown are enforced on the platform, not in the workflow.

For the trigger, add a gtm-api Trigger node, pick an event, and activate the workflow: it registers the webhook on activation and removes it on deactivation. Each execution receives the delivery envelope as its item, so {{ $json.type }} is the event name and the event body is under {{ $json.payload }}. The n8n instance has to be reachable over public https. The platform refuses a non-public target URL.

Links

• API docs: docs.gtm-api.com
• What the platform does and the safety model: gtm-api.com
• MCP server for AI agents: github.com/gtm-api/linkedin-mcp

License

MIT