What is the Model Context Protocol?
The Model Context Protocol (MCP) is an open standard, created by Anthropic in 2024, for connecting AI agents to external data and tools. Think of it as a USB-C port for AI: one standard interface that works with any compatible agent or tool. Instead of bespoke integrations, an agent connects to a server, discovers the available tools, and calls them with natural language.
Unlike a REST API — where a developer reads docs and writes code for each endpoint — the protocol exposes a self-describing tool manifest the agent reads automatically. That is what makes it ideal for lead generation: the agent figures out how to search and verify leads on its own.
2024
Created by Anthropic
9,652
servers in the official registry (May 2026)
97M+
monthly protocol SDK downloads
USB-C for AI
one standard interface for any agent
Sources: Model Context Protocol and the official protocol registry.
How Leadloadz's server works
- Server endpoint
https://leadloadz.com/api/mcp - npm package
@leadloadz/mcp-server - Auth
Token-based (create in your dashboard) - Transport
HTTP (JSON-RPC 2.0) or stdio
The server ships three tools — search_leads, verify_email, and get_user_stats — all returning structured JSON-RPC 2.0 responses.
Agent tools reference
search_leadsFind contacts by criteria.
Params: query (string), limit (number, max 100)
Returns: Array of leads with name, title, company, email, verified status
verify_emailValidate email deliverability.
Params: email (string)
Returns: Verification status, confidence score, deliverability
get_user_statsCheck quota usage.
Params: none (uses auth token)
Returns: searches used, searches remaining, plan type
Example request and response:
// search_leads request
{
"name": "search_leads",
"arguments": {
"query": "CTOs at Series A SaaS companies in London",
"limit": 20
}
}
// response
{
"results": [
{
"name": "Jane Smith",
"title": "CTO",
"company": "TechCorp",
"email": "jane@techcorp.com",
"verified": true,
"location": "London, UK"
}
]
}Setting up Claude Desktop with Leadloadz
- 1Create an account and an API token.
- 2Open Claude Desktop → Settings → Developer → Edit Config.
- 3Add the block below to claude_desktop_config.json.
- 4Restart Claude Desktop.
- 5Ask Claude: “Find CTOs at Series A SaaS companies in London.”
{
"mcpServers": {
"leadloadz": {
"command": "npx",
"args": ["-y", "@leadloadz/mcp-server"],
"env": {
"LEADLOADZ_API_TOKEN": "your_token_here"
}
}
}
}Setting up Cursor with Leadloadz
- 1Create an account and an API token.
- 2Open Cursor → Settings → MCP → Add Server.
- 3Add the Leadloadz server configuration (same command/args/env as Claude).
- 4Use it from Chat (Cmd/Ctrl+L) or Composer: “Find leads for my outreach campaign.”
MCP vs REST API: which to use?
| Feature | MCP | REST API |
|---|---|---|
| Setup | Config file | Code integration |
| Auth | Token in env | Bearer header |
| Interface | Natural language | HTTP requests |
| Best for | Claude/Cursor users | Custom applications |
| Tool discovery | Automatic | Manual endpoint docs |
Recommendation: use the protocol if you work in Claude or Cursor. Use the REST API if you are building a custom application.
Example agent workflows
“Find 20 verified VP Engineering contacts at fintechs in NYC.”
“Verify these 50 email addresses before my campaign.”
“How many searches do I have left this month?”
“Find decision-makers at companies similar to Stripe.”
Frequently asked questions
What is MCP lead generation?
It connects AI agents to a lead-data source using the Model Context Protocol — an open standard for exposing tools to LLM agents. Your agent calls a search_leads tool and receives structured, verified B2B contacts it can act on, with no manual exporting.
How do I connect Claude to Leadloadz?
Create an API token in your dashboard, open Claude Desktop → Settings → Developer → Edit Config, add the Leadloadz MCP server block to claude_desktop_config.json, restart Claude, and ask it to find leads. Setup takes about two minutes.
What tools does Leadloadz provide?
Three: search_leads (find contacts by criteria), verify_email (validate deliverability in real time), and get_user_stats (check your quota and plan). All speak JSON-RPC 2.0 and authenticate with a bearer token.
Is the Leadloadz server free?
Yes — the free tier includes 10 searches/month with no credit card, plus one API token and full server access. Paid plans add more volume and email verification.
How many leads can I search per month?
It depends on your plan: Free includes 10 searches/month, Starter ($29/mo) includes 250, and Pro ($79/mo) includes 1,000. Each search can return multiple verified contacts.
What is the difference between MCP and REST API?
The protocol is config-driven and works through natural language inside agents like Claude and Cursor, with automatic tool discovery. The REST API is code-driven with manual HTTP requests and endpoint docs — better for custom applications. Both hit the same infrastructure.
Does the server work with Cursor?
Yes. Add the Leadloadz server in Cursor → Settings → MCP, then use it from Chat or Composer to find and verify leads without leaving your editor.
How do I verify emails with the server?
Call the verify_email tool with an email address (or ask your agent to verify a list). It returns a deliverability status and confidence score using a real-time 5-layer MX/SMTP check.