Skip to main content
The Lusha MCP server implements the Model Context Protocol - a standard that lets AI assistants and agents call external tools directly. It’s generally available today at https://mcp.lusha.com. Once connected, an MCP-compatible client can enrich contacts, search for prospects, and retrieve company data using natural language - without you writing a single API call.
The Lusha MCP server has moved beyond its original beta. Some connection details below (client list, exact tool set) can change as Lusha adds tools - always check the tools menu in your client for the current list.
Starter prompt included. Every new Lusha MCP connection ships with a built-in starter prompt on the client’s consent screen - copy it in one click to run a real first query (find a person, verify they still work there, and pull their title and verified contact details) as soon as you connect. This is available across every MCP client below, not just Claude.
22
MCP tools
Enrichment, prospecting, signals, lookalikes, recommendations, and account tools.
7
Supported clients
Claude, ChatGPT, Codex, VS Code, Cursor, Gemini CLI, and n8n - plus any other MCP client.
2
Auth methods
OAuth for chat clients, API key header for everything else.

​
What you can do with Lusha MCP

  • Ask an AI assistant to “enrich this contact” and get back verified emails and phone numbers
  • Instruct an agent to “find 20 VP of Engineering contacts at Series B companies in New York”
  • Find companies or contacts that look like your best customers (lookalikes)
  • Check recent hiring, headcount, or news signals for an account before you reach out
  • Trigger a prospect list build from inside a Claude, ChatGPT, Cursor, or Gemini CLI session

​
How it works

Your AI client
Claude, ChatGPT, Cursor, VS Code, Gemini CLI, n8n…
→
Lusha MCP server
mcp.lusha.com - translates tool calls into API requests
→
Lusha API
Enriched data returned to your client
The MCP server translates natural-language tool calls into authenticated Lusha API requests and returns structured results that the AI can reason over or pass downstream. Two connection methods are supported:

OAuth

For Claude, ChatGPT, and Codex - connect your Lusha account through the client’s own connector directory. No API key handling required on your side.

API key

For VS Code (GitHub Copilot), Cursor, Gemini CLI, n8n, and any other MCP client - pass your key via an x-api-key header.
MCP tool calls use the same Lusha API key and credit pool as the REST API - see MCP setup for connecting each client.

​
Next steps

Setup

Connect the Lusha MCP server to Claude, ChatGPT, Cursor, VS Code, Gemini CLI, or n8n.

Available tools

Browse every tool the MCP server exposes.

Examples

Prompt examples for common enrichment and prospecting tasks.

Get API Key

You need a Lusha API key to use the MCP server with API-key-based clients.