Linkedin MCP server
{
"mcpServers": {
"gtm-api": {
"type": "http",
"url": "https://mcp.gtm-api.com/mcp"
}
}
}{
"mcpServers": {
"gtm-api": {
"type": "http",
"url": "https://mcp.gtm-api.com/mcp"
}
}
}{
"mcpServers": {
"gtm-api": {
"type": "http",
"url": "https://mcp.gtm-api.com/mcp"
}
}
}{
"servers": {
"gtm-api": {
"type": "http",
"url": "https://mcp.gtm-api.com/mcp"
}
}
}{
"mcpServers": {
"gtm-api": {
"type": "http",
"url": "https://mcp.gtm-api.com/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @gtm-api/linkedin-mcpversion 1.1.0
What it does
A managed LinkedIn MCP server that exposes 160+ LinkedIn actions (search, connect, message, enrich) through three MCP tools. It runs LinkedIn accounts in isolated cloud infrastructure with safety limits enforced server-side, and works with Claude, ChatGPT, Cursor, and other MCP-compatible clients.
- Accept connection invitations from specific profiles and send personalized messages to new connections automatically.
- Search for LinkedIn profiles matching criteria and enrich them with account data without touching the LinkedIn API.
- Build workflows that manage connection outreach, screening, and follow-up across multiple LinkedIn accounts through agents.
Tools this server exposes
| Tool | Does |
|---|---|
| list_toolsets | List the available toolsets (domains) on the server, such as linkedin.messaging or id.billing. |
| get_toolset_tools | List the tools in a toolset with their descriptions, safety flags, and parameters. |
| call_tool | Execute a specific LinkedIn action from a toolset. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.