The Lusha MCP Server brings enterprise-grade B2B contact and company data directly to your AI conversations. Built on Lusha's database of 100+ million business contacts and 60+ million companies, it enables AI assistants to find, verify, enrich, and prospect contacts and companies through natural language - without writing API calls or managing technical integrations.
MCP (Model Context Protocol) is an open standard that allows AI assistants to securely connect to external data sources and tools. Lusha's MCP Server is compatible with Claude, ChatGPT, Cursor, VS Code, and any other MCP-supported AI client.
Features
Contact Data Enrichment - access verified email addresses, phone numbers, and professional profiles using email, LinkedIn URL, or name + company.
Company Data Enrichment - retrieve firmographics, employee counts, industry data, funding information, and business insights.
Bulk Processing - enrich up to 100 contacts or companies in a single request for high-volume workflows.
Prospecting - search Lusha's full database using structured filters (job title, seniority, location, company size, industry, and more) to build targeted prospect lists.
Requirements
A valid Lusha API key - Admin or Manager role required
A Lusha plan with sufficient credits
An MCP-compatible AI client (Claude, ChatGPT, Cursor, VS Code, and others)
Your API key and personal MCP URL are both available in API Hub (left sidebar). Click Copy API key in the top right to copy your key, or open the MCP tab to find your personal MCP URL and platform setup guides.
Setup
Option 1 - Claude Remote MCP
This is the easiest setup route. It works in Claude Web and Claude Desktop with no configuration files required. Best for non-technical users, quick pilots, and shared team setups.
Open Claude and go to Settings > Connectors.
Click Add custom connector.
Paste the Lusha Remote MCP URL: https://mcp.lusha.com.
Complete authentication via OAuth.
In any chat, click the + button at the bottom left of the chat, select Connectors, and enable Lusha to confirm the tools are active.
Option 2 - ChatGPT
The Lusha MCP Server supports ChatGPT. An Admin must configure the connector first; individual users can then enable it.
Admin setup:
In ChatGPT, go to Settings > Manage Workspace.
Click Apps & Connectors > Create.
Toggle on Dev Mode.
Enter the following: Name: Lusha MCP / URL: https://mcp.lusha.com / add your API authentication details.
Click Create, then toggle the connector ON.
User setup (after Admin has enabled the connector):
In ChatGPT, go to Settings > Apps & Connectors.
Find Lusha Connector and toggle Enable.
⚠️ Important — Lusha must be actively invoked in each ChatGPT conversation
Enabling the connector does not make Lusha active in every conversation automatically. You must use one of these two methods in every conversation:
Start from the Lusha Connector page: Find Lusha in the ChatGPT Marketplace and click Start chat.
Use @Lusha in the prompt bar: Type @Lusha before your request and select Lusha from the dropdown. For example: find 2 product marketing VPs in Lisbon with @Lusha
Option 3 - Cursor
Go to Cursor > Settings > Cursor Settings > MCP.
Add a Lusha entry with the command:
npx @lusha/mcp.Set your Lusha API key in the environment variables section.
Optional: enable Yolo mode to approve tool calls automatically.
Option 4 - n8n
You can connect Lusha to n8n using the MCP setup guide available on the MCP tab in API Hub.
Available tools
The Lusha MCP Server provides the following tools. For prospecting, always start with the prospecting_search_guide tool before using filters or search — it provides step-by-step workflow guidance and helps avoid unnecessary credit usage.
Category | Tool | Description |
Contact Data | contacts_search | Enrich contact information using email, LinkedIn URL, or name + company. Up to 100 contacts per request. |
Company Data | companies_search | Get comprehensive company data using domain or company name. Up to 100 companies per request. |
Prospecting - Guide | prospecting_search_guide | Start here. Step-by-step guidance for optimal prospecting workflows. Call this first before any prospecting tool. |
Prospecting - Companies | prospecting_company_filters | Get available filter options 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 and technographic data. May consume credits. |
Prospecting - Contacts | prospecting_contact_filters | Get available filter values for contact search. No credits charged. |
| prospecting_contact_search | Search for contacts using structured filters. |
| prospecting_contact_enrich | Enrich contacts by ID to reveal emails and phone numbers. Consumes credits. |
💡 Tip: Recommended prospecting workflow: prospecting_search_guide first → use filter tools to preview valid values (no credits) → search → enrich only the results you want.
Example prompts
"Can you get Yoni Tserruya's email from Lusha?"
"Find 15 senior directors of IT at 200-1,000 employee SaaS companies in the UK. Enrich the top 5 with work email and mobile."
"Find 10 contacts in Boston working at technology companies."
"Search for marketing managers at SaaS companies in New York."
"Get company information for Lusha, including employee count, industry, and location."
Credits and billing
Understanding exactly how credits are consumed helps you avoid surprises, especially when running high-volume MCP workflows.
Which tools consume credits?
Tool | Credits charged? | Cost |
prospecting_search_guide | No | Free |
prospecting_company_filters | No | Free |
prospecting_contact_filters | No | Free |
contacts_search | Yes | 1 credit per email · 5 credits per phone number |
companies_search | Yes | 1 credit per company profile |
prospecting_contact_search | Yes | 1 credit per 1–25 results returned |
prospecting_company_search | Yes | 1 credit per 1–25 results returned |
prospecting_contact_enrich | Yes | 1 credit per email · 5 credits per phone number |
prospecting_company_enrich | Yes | 1 credit per company enriched |
💡 Note: Each contact's data point is charged only once across all Lusha channels. If you've already revealed a contact's email through the platform, extension, or API, you won't be charged again for the same data point via MCP.
💡 Note: Failed lookups that return no results are not charged.
Example credit costs
Enrich 10 contacts with email and phone → up to 60 credits (10 × 1 for email + 10 × 5 for phone).
Search for 50 contacts using prospecting_contact_search → 2 credits (2 pages of up to 25 results each).
Run prospecting_contact_filters to preview seniority filter values → 0 credits.
How to track MCP credit usage
MCP credit consumption is logged alongside all other API usage. Monitor it in Lusha Analytics → API Usage tab, filtered by API key and date.
How to set a credit cap for MCP
Go to Account & Settings → API Hub.
Select the API key used for your MCP connection.
Set a Credit limit on that key. Once the limit is reached, MCP calls using that key will stop consuming credits until the limit is reset or raised.
Rate limits: 25 requests per second. For high-volume workflows, use bulk operations (up to 100 contacts or companies per request) to stay within limits efficiently.
Security
Store your API key in secure credentials - use your AI client's built-in credentials manager or OS keychain. Never paste it in plain text in a chat.
Limit tool permissions in Claude Connectors to the minimum needed for your workflow.
Rotate your API key periodically and remove unused connectors from your AI clients.
Troubleshooting
Lusha tools not appearing in Claude
If you've connected Lusha but don't see any Lusha tools available in your Claude conversation, work through the steps below in order.
Claude Web (Remote MCP)
Go to Claude Settings → Connectors and confirm the Lusha connector is listed and shows as connected.
In your chat, click the + button at the bottom left and select Connectors. Make sure Lusha is toggled on for this conversation.
If Lusha shows an error, click Reconnect and complete the OAuth flow again.
If Lusha is not listed at all, go to Settings → Connectors → Add custom connector, paste https://mcp.lusha.com, and complete setup.
Claude Desktop (config file)
Confirm the claude_desktop_config.json file exists in the correct location for your OS.
Open the file and verify your API key is correctly pasted — no extra spaces, quotes, or line breaks.
Fully quit and reopen Claude Desktop. Changes to the config file are only loaded on restart.
After restarting, look for the hammer icon (🔨) in the chat input — this confirms MCP tools are loaded.
💡 Note: Node.js must be installed for the npx @lusha/mcp command to work. Download it from nodejs.org if you haven't already, then restart Claude Desktop.
"API Key Not Found" error
Verify your API key is correctly set in the configuration file or environment variables.
Confirm API access is enabled on your Lusha plan (Admin or Manager role required).
Check that the key has not been reset since you set it up.
"I can't access Lusha" error in ChatGPT
This means Lusha was not actively invoked in this conversation. Use one of the two methods below:
Go to the ChatGPT Marketplace, find Lusha, and click Start chat.
Or type @Lusha in the prompt bar and select Lusha from the dropdown before sending your message.
"No Results Found"
Try a different identifier - email, LinkedIn URL, or full name + company.
Verify the contact or company exists in Lusha's database.
Check for typos in the provided information.
Partial or no results returned
Credits follow standard API rules - no charge on failed lookups. If your account has geographic restrictions, some fields may return as null for compliance reasons.
Got a feature request? Submit it here!