The Lusha MCP Server connects Lusha's B2B contact and company data directly to your AI workflows. Use it to find, enrich, and prospect through natural language - no manual API calls required.
Lusha MCP supports two authentication methods. Which one you use depends on your client:
| Method | Description | Supported clients |
|---|---|---|
| OAuth | One-click authentication via Lusha's official connector | Claude, ChatGPT |
| API Key | Pass your Lusha API key via the x-api-key request header | VS Code, Cursor, n8n, Gemini CLI, and any other client |
Connect via Lusha's official listing in the Claude connector directory.
- Go to claude.ai/directory/connectors/lusha
- Click Add to Claude
- Complete OAuth authentication
- Open any chat, click the tools menu, and confirm Lusha tools are listed
Connect via Lusha's official listing in the ChatGPT app directory.
- Go to chatgpt.com/apps?q=lusha
- Click Connect
- Complete OAuth authentication
- Open any chat, click the tools menu, and confirm Lusha tools are listed
For clients not listed above, connect using your Lusha API key.
Get your API key: Sign in to Lusha and go to API → Manage API Keys.
MCP Server URL: https://mcp.lusha.com
Requires VS Code 1.99 or higher.
Option A: Settings UI
- Open the Command Palette (
Cmd+Shift+P/Ctrl+Shift+P) - Run MCP: Add Server
- Select HTTP, enter
https://mcp.lusha.com, and name itlusha - Add
x-api-keyas the header name and your API key as the value when prompted
Option B: settings.json
{
"mcp": {
"servers": {
"lusha": {
"type": "http",
"url": "https://mcp.lusha.com",
"headers": {
"x-api-key": "YOUR_LUSHA_API_KEY"
}
}
}
}
}To use Lusha tools, open GitHub Copilot Chat, switch to Agent mode, and enable Lusha from the Tools menu.
Open (or create) your Cursor MCP config:
- macOS/Linux:
~/.cursor/mcp.json - Windows:
%APPDATA%\Cursor\mcp.json
- macOS/Linux:
Add the following:
{
"mcpServers": {
"lusha": {
"url": "https://mcp.lusha.com",
"headers": {
"x-api-key": "YOUR_LUSHA_API_KEY"
}
}
}
}- Restart Cursor - Lusha tools will be available automatically in Chat and Composer.
Best for developers using Gemini CLI as their AI-powered terminal who need B2B data for prospecting, enrichment, or research without leaving the command line.
Option A: Extension install (recommended)
gemini extensions install lusha-oss/lusha-gemini-extensionThen set your API key as an environment variable:
export LUSHA_API_KEY=YOUR_LUSHA_API_KEYAdd the export to your shell profile (~/.bashrc, ~/.zshrc, etc.) to persist it across sessions.
Option B: Manual settings.json (Streamable HTTP)
{
"mcpServers": {
"lusha": {
"httpUrl": "https://mcp.lusha.com",
"headers": {
"Authorization": "Bearer YOUR_LUSHA_API_KEY"
}
}
}
}Option C: Manual settings.json (npx/stdio)
{
"mcpServers": {
"lusha": {
"command": "npx",
"args": ["@lusha-org/mcp@latest"],
"env": {
"LUSHA_API_KEY": "YOUR_LUSHA_API_KEY"
}
}
}
}Using Lusha in Gemini CLI
Once connected, you can ask Lusha questions in natural language directly in your terminal:
- "Find the VP of Engineering at Stripe"
- "Enrich this contact: jane.doe@acme.com"
- "Find 20 Series B SaaS companies in New York hiring in sales"
If you installed via the extension, you also have access to slash commands:
/lusha:prospect- run a prospecting search/lusha:enrich- enrich a contact or company
Requires n8n 1.88.0 or higher.
- In your workflow, add an MCP Client Tool node
- Set Connection Type to HTTP Streamable
- Enter
https://mcp.lusha.comas the URL - Under Authentication, select Header Auth and set:
- Header name:
x-api-key - Value:
YOUR_LUSHA_API_KEY
- Header name:
- Under Tools to Use, select All Tools or choose specific ones
- Connect the node to your AI Agent node and activate the workflow
Store your API key using n8n's built-in Credentials system rather than pasting it directly into node fields.
Any MCP-compatible client can connect to the Lusha MCP Server using the same pattern:
- URL:
https://mcp.lusha.com - Header:
x-api-key: YOUR_LUSHA_API_KEY
Refer to your client's MCP documentation for where to configure the server URL and request headers.
| Tool | Description |
|---|---|
contacts_search | Enrich contact data using email, LinkedIn URL, or name + company |
companies_search | Retrieve company data using domain or company name |
prospecting_contact_filters | Get available filter values for contact prospecting (no credits charged) |
prospecting_contact_search | Search contacts by job title, seniority, location, department, and more |
prospecting_contact_enrich | Enrich contacts by ID to reveal emails and phones (credits consumed) |
prospecting_company_filters | Get available filter values for company prospecting (no credits charged) |
prospecting_company_search | Search companies by firmographic and technographic filters |
prospecting_company_enrich | Enrich companies by ID for detailed firmographic data |
lookalike_contacts | Find contacts similar to a seed contact |
lookalike_companies | Find companies similar to a seed company |
signals_contact_filters | Get available filter values for contact signals (no credits charged) |
signals_contacts_search | Resolve contacts and return recent activity signals |
signals_contacts_get | Get signals for known contacts by ID |
signals_company_filters | Get available filter values for company signals (no credits charged) |
signals_companies_search | Resolve companies and return recent activity signals |
signals_companies_get | Get signals for known companies by ID |
account_usage | Check credit balance, rate-limit status, and plan info |
Available tools may vary depending on the client used to connect.
| Issue | Resolution |
|---|---|
| Tools not showing in Claude | Confirm the connector is added and you have permission to use tools. Restart the app. |
| Tools not showing in ChatGPT | Confirm the connector is connected and enabled in the tools menu. Restart the app. |
| Tools not showing in VS Code | Make sure you're in Agent mode in Copilot Chat. Verify mcp.servers is correctly formatted in settings.json. |
| Tools not showing in Cursor | Verify mcp.json is in the correct path and is valid JSON. Restart Cursor fully. |
| Tools not showing in n8n | Confirm the MCP Client Tool node is connected to your AI Agent node and the workflow is active. Check the URL and header format. |
| Tools not showing in Gemini CLI | Verify settings.json is correctly formatted. Run /mcp reload to refresh. Restart Gemini CLI. |
| Auth errors - wrong header | For custom MCP clients, the header must be x-api-key (lowercase). Using Authorization: Bearer or any other casing will fail. |
| Auth errors - expired key | API keys can be revoked or expired. Verify your key is active in API Hub → Manage API Keys and generate a new one if needed. |
| Partial or no results | Credits follow the same rules as the Public API - failed lookups are not charged. Check your credit balance in the Lusha admin panel. |
| OAuth not completing | If the OAuth flow stalls or errors, try disconnecting and reconnecting the connector from the directory listing. Make sure you are signed in to Lusha before starting the flow. |
- Store API keys in secure credentials (OS keychain, password manager, or your editor's secrets manager) - avoid hardcoding them in shared config files
- In n8n, use the built-in Credentials system rather than pasting keys directly into node fields
- Use environment variables for team setups instead of committing keys to version control
- Rotate API keys periodically and remove unused connectors
- Limit tool permissions to the minimum needed for your use case