gtmapi/linkedin-mcp repository overview

⁠gtm-api: LinkedIn MCP Server

Managed LinkedIn MCP server for AI agents. This image is a thin launcher: it bridges a stdio MCP client (Claude Desktop, Cursor, anything that spawns a command) to the hosted endpoint at https://mcp.gtm-api.com/mcp with mcp-remote. No server logic runs in the container.

⁠Quickstart

1. Get an API key at app.gtm-api.com⁠ (free plan for one account, no card) and connect a LinkedIn account.
2. Add the server to your MCP client config:

{
  "mcpServers": {
    "gtm-api": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "-e", "GTM_API_KEY", "gtmapi/linkedin-mcp"],
      "env": { "GTM_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Connecting over HTTP directly, without Docker? Point the client at https://mcp.gtm-api.com/mcp with an Authorization: Bearer header instead.

⁠What the agent can do

250+ typed tools over the Model Context Protocol, 175+ of them on LinkedIn: search people, companies and posts, send connection requests, message, InMail, enrich profiles, track content, manage the cloud browsers the accounts run in. Every outward action returns a preview and requires confirmation, and server-side daily limits are checked before dispatch.

⁠Is it safe for my LinkedIn account?

Each connected account runs in its own cloud browser with a residential proxy in its country. New accounts warm up on a ramp, and Smart Limits set each account's daily caps per action from six live health signals, enforced on the server. 20,000+ LinkedIn accounts run on this stack, and under 1% of them have ever been restricted. LinkedIn's User Agreement does not permit third-party automation, so every tool in this space carries risk. This is not legal advice.

⁠Links

• How a LinkedIn MCP server works: gtm-api.com/linkedin-mcp-server⁠
• Source and full docs: github.com/gtm-api/linkedin-mcp⁠
• MCP registry name: com.gtm-api/linkedin-mcp