How to use Snov.io MCP with your AI Assistant

This guide shows you how to connect your Snov.io account to an AI assistant and run lead generation tasks through natural language.

What is MCP

MCP is an open standard that lets AI assistants connect to external tools and services securely. It gives your AI direct access to your Snov.io account without tab switching or API calls.

What you can do with it

Snov.io MCP lets you connect your Snov.io account to AI assistants like Claude, ChatGPT, Cursor, and others. 

Once connected, your AI assistant can perform real actions in your Snov.io account: search for leads, verify emails, manage your CRM, and run LinkedIn outreach — just tell your AI what you need to do.

What you need to use it

Before you start, make sure you have:

• An active Snov.io account (any plan)
• An AI assistant that supports MCP: Claude Desktop, ChatGPT, Cursor, Windsurf, or another MCP-compatible client

What Snov.io MCP can do

Snov.io MCP includes more than 100 actions. Here is the list of tasks currently available in each.

Prospect search and enrichment

• Search the Snov.io database by AI-powered filters (job title, industry, location, company size, revenue, and more) and save to lists
• Find contacts by company domain (start a search, get results, find contacts by domain)
• Find and verify individual emails by name and company
• Pull a prospect profile by email, ID, or list
• Search LinkedIn profiles by filters or URL and save them as prospects
• Verify emails individually or in bulk
• List all prospects in a list
• Add, update, and remove prospects in lists
• Create, rename, delete, restore, and merge prospect lists
• Organize lists into folders: list all folders, create new, move a list from one folder to another, delete folder
• Export lists with custom field selection

CRM

Pipelines

• List, create, edit, reorder, and delete pipelines
• Manage pipeline statuses

Deals

• List, create, update, search, and delete deals
• Move deals between stages
• Mark deals as won or lost with reason tracking
• Add participants to deals
• View full deal timeline and activity history
• Manage deal currencies and colors

Notes

• Add, edit, and delete notes on deals

Loss reasons

• List, create, update, and delete loss reasons

Tasks

• List and filter tasks by status, owner, or related deal
• Create and update calendar tasks: calls, meetings, follow-ups
• Mark tasks as done
• Link tasks to deals and contacts

LinkedIn account setup

• Connect and manage multiple LinkedIn accounts
• Configure daily action limits for safe automation
• Set up warm-up mode for new accounts
• Manage proxies and server locations
• Assign and manage team slots
• Check trial access and usage summary

LinkedIn actions

• Send connection invites, messages, and InMails
• Visit profiles, like posts, endorse skills
• Check connection level and invite status
• View and sync chat history with prospects
• Parse LinkedIn network and export contacts as CSV
• Convert LinkedIn profiles into Snov.io prospects
• Save prospects from network or pending connections

Webhooks

• List, create, update, and delete webhooks

How to set it up

Step 1. Log in to your Snov.io account and keep it open

Step 2. Add Snov.io MCP to your AI assistant

The exact steps depend on which AI assistant you use. Below is the general process:

Claude Desktop:

1. Open Claude Desktop → Settings → Connectors →  Add custom connector

2. In the field Name, type Snovio MCP.

In the field Remote MCP Server URL, add https://mcp.snov.io/mcp

3. Click Add and then Connect to Snov.io.4. Next, Approve Claude's access to Snov.io.

Now, you're all set:

ChatGPT, Cursor, or other clients:

Follow the MCP server setup instructions for your specific client. You’ll need the same server URL and credentials from Step 1.

Step 3. Verify the connection

Once connected, run a simple command to confirm everything works: choose from our available MCP actions. Your AI assistant should return a response from your Snov.io account, confirming the connection is active.

How to write prompts for your AI Agent 

Be specific. Instead of "Find leads", say "Find VP of Sales at SaaS companies in the UK with 100-500 employees."

Chain actions. Your AI can run multi-step workflows in one conversation: search → verify → save → create deals without switching tools.

Use plain language. You do not need to remember exact field names or API parameters. Describe what you need in plain words and the AI handles the rest.

Check async results. Some actions run in the background, like LinkedIn network parsing and bulk verification, for example. Ask your AI to check their status after a few minutes.

Here are ready-to-use workflows you can run through chat:

Workflow	Sample prompts
Lead generation	"Search the Snov.io database for CTOs at fintech companies in Germany with 50 to 200 employees." "Find all marketing directors in SaaS companies in the US." "Verify the emails from those results." "Save the verified contacts to a new list called 'Fintech CTOs — Germany'." "Create a deal for each lead in my Sales Pipeline."
LinkedIn outreach	"Show me all connected LinkedIn accounts and their safety limits." "Send a connection invite to linkedin.com/in/example-profile with the note: Hi, I'd love to connect!" "Check the invite status for that profile." "Once they accept, send them a follow-up message."
CRM management	"List all deals in my Enterprise pipeline that are in the Negotiation stage." "Move the Acme Corp deal to Won." "Add a note to the deal: Contract signed July 2026." "Create a follow-up task for next Monday." "Schedule a follow-up call with the Acme Corp deal for tomorrow at 3 PM."

FAQs

Which AI assistants are supported? Any AI assistant that supports the MCP standard, including Claude Desktop, ChatGPT, Cursor, Windsurf, and more.

Is Snov.io MCP available on all plans? Yes, it is available on all Snov.io plans.

Do I need to know how to code? No. You control everything through natural language in your AI assistant.

Are my credentials safe? Yes. Snov.io MCP uses secure login/OAuth authentication. You never share raw API keys.

What if my AI assistant does not support MCP yet? MCP is an emerging standard and more AI tools are adding support regularly. Check your assistant's documentation, or use one of the supported clients listed above.