Use Emailyze in Claude and Cursor via MCP

Emailyze provides a hosted remote MCP server at https://api.emailyze.xyz/mcp/. No installation, no Python environment, no local process to manage. You add a URL and an API key to your AI client's config, restart it, and email verification becomes a native tool.

This guide covers setup for Claude Desktop and Cursor. Both support remote MCP servers using the same config pattern.


What You Need


Claude Desktop Setup

1. Open the config file

If the file doesn't exist yet, create it.

2. Add the Emailyze server

{
  "mcpServers": {
    "emailyze": {
      "url": "https://api.emailyze.xyz/mcp/",
      "headers": {
        "x-api-key": "YOUR_EMAILYZE_API_KEY"
      }
    }
  }
}

Replace YOUR_EMAILYZE_API_KEY with your actual key. If you have other MCP servers already configured, add the emailyze block inside the existing mcpServers object.

3. Restart Claude Desktop

Fully quit the app (⌘Q on macOS, not just closing the window) and relaunch. You should see a hammer icon in the chat input area. Click it to confirm check_email and check_emails_batch are listed.

4. Try it

> Is test@guerrillamail.com a real email address?

Claude will call check_email automatically and respond with the risk score, provider type, and a plain-language recommendation.


Cursor Setup

Cursor supports remote MCP servers in its settings file.

1. Open Cursor MCP settings

Go to Settings → Cursor Settings → MCP (or open the Command Palette and search "MCP"). Alternatively, edit the settings file directly:

2. Add the Emailyze server

{
  "mcpServers": {
    "emailyze": {
      "url": "https://api.emailyze.xyz/mcp/",
      "headers": {
        "x-api-key": "YOUR_EMAILYZE_API_KEY"
      }
    }
  }
}

3. Restart Cursor

Restart Cursor. In the MCP settings panel you should see the Emailyze server listed as connected with the check_email and check_emails_batch tools available.

4. Use it in Cursor Chat

Open Cursor Chat (Cmd+L) and try:

> Check if signup@mailnull.com is a disposable email and whether I should allow it in my user registration flow.

Cursor will call check_email and explain the result in the context of your codebase.


Available Tools

`check_email`

Check a single email address or domain.

Parameters:

Returns: is_disposable, risk_score (0–100), provider_type, mx_valid, mx_provider, catch_all, domain, syntax, sources, and more.

`check_emails_batch`

Check up to 1,000 email addresses in a single call.

Parameters:

Returns: Array of results in the same order as input.


Example Prompts

Single check: > Is noreply@tempmail.com a disposable address? Should I block it?

Batch check: > I'm reviewing new signups from this week. Check these emails and flag any I should be concerned about: alice@gmail.com, bob@tempmail.com, carol@proton.me, dave@guerrillamail.com

Code-context check (Cursor): > I have this user registration handler open. Check whether the email field needs stronger validation — use the Emailyze tool to test a few common disposable domains and tell me what risk scores come back.


How MCP Servers Are Discovered

When you add a URL to your MCP config, the client fetches that endpoint and asks it to list available tools. The protocol exchange looks like this:

Client → POST /mcp/   {"method": "tools/list"}
Server → {"tools": [{"name": "check_email", ...}, ...]}

The client caches the tool list and shows you what's available. When you ask a question that seems relevant to a tool, the model decides whether to call it — and calls it automatically if it does.

Three ways AI clients discover MCP servers without manual config:

1. /llms.txt — A plain-text file at https://api.emailyze.xyz/llms.txt that describes the API in natural language including MCP config snippets. Some AI agents fetch this automatically when building context about a domain.

2. OpenAPI spec — Machine-readable API definition at https://api.emailyze.xyz/v1/openapi.json. Agent frameworks like LangChain, CrewAI, and AutoGen can read OpenAPI specs and auto-generate tool wrappers from them.

3. /.well-known/x402.json — The x402 discovery manifest at https://api.emailyze.xyz/.well-known/x402.json lists payment-gated endpoints. Agent frameworks that support x402 can find the email check endpoint from this manifest without any manual configuration.

For most users, manual config (the steps above) is the right approach — it's two lines and takes two minutes. Discovery protocols matter more for autonomous agents that programmatically find and register tools at runtime.


Troubleshooting

Tools not appearing after restart:

SSL / certificate error:

401 Unauthorized:

Cursor shows "No server info found":


Next Steps