Skip to main content

MCP Server Beta

The LeadMagic MCP server lets AI assistants call LeadMagic directly. Ask your AI to validate contacts, find emails, enrich companies, submit bulk jobs, check account analytics, or build an account brief - the assistant uses the right LeadMagic tool and returns the result in context.
Beta — The MCP server is fully functional but still in beta. Tools, parameters, and behavior may change. We’d love your feedback — reach out at support@leadmagic.io or via the Help Center.
What is MCP? The Model Context Protocol is an open standard that lets AI assistants use external tools. LeadMagic’s MCP server exposes contact, company, jobs, bulk, analytics, and workflow tools that compatible AI clients can call.

What You Can Do

Ask your AI assistant questions like:
  • “Find the work email for Alex Rivera at Example Robotics”
  • “Validate these 50 email addresses before my campaign”
  • “What’s the tech stack at example.com?”
  • “Find and validate the CEO, CRO, and VP Sales at Example Robotics”
  • “Open the upload widget and enrich this company list”
  • “How much did I use today and am I close to my limits?”
  • “Give me account intel on Example Robotics: funding, competitors, jobs, and tech stack”
The AI calls the right LeadMagic tool automatically - no code, no API calls, no context switching.

What The MCP Can Do

ToolWhat It Does
validate_work_emailVerify deliverability before outreach
find_work_emailName + company/domain to work email
find_mobile_numberProfile/email to direct mobile
linkedin_profile_to_work_emailProfile URL to work email
email_to_b2b_profileEmail to B2B profile URL
search_profileProfile URL to professional profile
find_personal_emailProfile URL to personal email
detect_job_changeCheck whether a contact moved jobs
enrich_contactComposite dossier across validation, profile, personal email, mobile, and job change
See the full tool reference for tool categories, parameters, examples, cost behavior, resources, and prompts.

Supported Clients

Works with any MCP client that supports Streamable HTTP transport — 13 clients and counting.

Cursor

Native Streamable HTTP support.

VS Code / Copilot

Secure input variables supported.

Claude Desktop

Via mcp-remote proxy.

Claude Code

One CLI command to add.

Windsurf

Native Streamable HTTP.

Zed

Via mcp-remote proxy.

Cline

Remote URL with OAuth.

Roo Code

Project or global config.

OpenCode

Remote type with OAuth.

Gemini CLI

CLI or settings.json.

JetBrains IDEs

IntelliJ, WebStorm, PyCharm, etc.

Amazon Q

HTTP transport in IDE or CLI.

Copilot Agent

GitHub coding agent.

How It Works

  1. You add the LeadMagic MCP server URL to your AI client
  2. Your client opens the LeadMagic OAuth sign-in flow
  3. The AI discovers the available LeadMagic tools automatically
  4. When you ask a question, the AI calls the right tool with the right parameters
  5. LeadMagic returns structured data, and the AI presents it in context
For bulk files, the MCP can open an upload widget, map columns, submit jobs, poll progress, inspect failures, and fetch completed rows. For account operations, it can estimate credit usage, check credits, and read analytics before running expensive work. The server runs at https://mcp.leadmagic.io/mcp using the Streamable HTTP transport. Claude remote connectors should use https://mcp.leadmagic.io.

Get Started

Setup Guide

Step-by-step setup for Cursor, VS Code, Claude, and Windsurf.

Tool Reference

All MCP tools with parameters, cost notes, resources, and prompts.

Authentication

OAuth setup and troubleshooting.

Troubleshooting

Common issues and how to fix them.