DomainScan MCP Quickstart (Model Context Protocol)

MCP (Model Context Protocol) lets AI assistants call external tools mid-conversation. With the DomainScan MCP server, Claude, Cursor, and other MCP-compatible clients can run domain lookups, DNS queries, SSL checks, blacklist scans, and email authentication analysis — directly, without leaving the conversation.

What You Can Do

Once connected, you can ask your AI assistant things like:

• “Check if api.example.com has SSL properly configured” → runs SSL check, explains results
• “Why is email from startup.io going to spam?” → probes SPF, DKIM, DMARC, MX, blacklists in one turn
• “Who owns competitor-domain.com and when does it expire?” → WHOIS lookup with parsed output
• “Is this IP address 185.220.101.34 a known threat?” → IP reputation + geolocation + blacklist check
• “Check DNS propagation for the A record I just changed on staging.example.com” → propagation check from global resolvers

The AI calls the right tools, chains multiple lookups if needed, and explains the results in plain language.

Prerequisites

• A DomainScan account and API key (get one here)
• An MCP-compatible AI client (Claude Desktop, Claude Code, Cursor, etc.)
• Node.js 18+ (for Claude Desktop/Code setup)

Installation

Option 1: npx (No Install)

The fastest way — no global install needed:

npx @domainscan/mcp-server --api-key YOUR_API_KEY

Test it runs correctly before adding to your client config.

Option 2: Global Install

npm install -g @domainscan/mcp-server
domainscan-mcp --api-key YOUR_API_KEY

Option 3: Clone and Run

git clone https://github.com/domainscan/mcp-server
cd mcp-server
npm install
npm start -- --api-key YOUR_API_KEY

Configuration

Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "domainscan": {
      "command": "npx",
      "args": ["@domainscan/mcp-server"],
      "env": {
        "DOMAINSCAN_API_KEY": "your_api_key_here"
      }
    }
  }
}

Restart Claude Desktop. You’ll see a hammer icon (🔨) in the input bar — click it to see available DomainScan tools.

Claude Code (CLI)

claude mcp add domainscan \
  --command "npx @domainscan/mcp-server" \
  --env "DOMAINSCAN_API_KEY=your_api_key_here"

Or add to your project’s .claude/settings.json:

{
  "mcpServers": {
    "domainscan": {
      "command": "npx",
      "args": ["@domainscan/mcp-server"],
      "env": {
        "DOMAINSCAN_API_KEY": "your_api_key_here"
      }
    }
  }
}

Cursor IDE

Open Cursor Settings → MCP → Add Server:

{
  "name": "domainscan",
  "command": "npx @domainscan/mcp-server",
  "env": {
    "DOMAINSCAN_API_KEY": "your_api_key_here"
  }
}

Environment Variable (Recommended for Security)

Instead of putting your API key directly in config files (which may be committed to git), use an environment variable:

export DOMAINSCAN_API_KEY=your_api_key_here

Then reference it in your config:

{
  "env": {
    "DOMAINSCAN_API_KEY": "${DOMAINSCAN_API_KEY}"
  }
}

Verify the Connection

After configuring, ask your AI assistant:

“Use DomainScan to check if google.com has DMARC configured”

The assistant should call the domainscan_dmarc_check tool, receive the result, and explain it. If it responds from memory without calling a tool, the MCP server isn’t connected.

In Claude Desktop, you can also click the 🔨 icon to see the full list of available tools.

Available Tools

The DomainScan MCP server exposes these tool groups:

Domain Tools

Network & IP Tools

Security Tools

Example: Full Email Deliverability Investigation

Prompt:

“Why is email from newsletter.company.com going to spam? Check everything.”

What happens behind the scenes:

1. domainscan_email_auth → checks SPF, DKIM, DMARC, BIMI, MTA-STS
2. domainscan_ip_blacklist → checks the MX server IPs against blacklists
3. domainscan_trust_score → checks overall domain reputation
4. Claude synthesizes all results into a plain-English diagnosis with specific fixes

That chain of lookups — which would take 5+ minutes manually — completes in seconds, and the AI explains exactly what to do.

Example: Pre-Launch Domain Checklist

Prompt:

“We’re launching newproduct.io next week. Run a full pre-launch check — DNS, SSL, email auth, blacklists, security headers.”

Result: a structured report covering everything that needs to be in place before launch, with specific issues and remediation steps.

Using Environment Variables in Claude Code

In Claude Code, you can reference the MCP server in .claude/settings.json at your project root. This makes DomainScan available automatically for any domain-related work in that project without a global install:

{
  "mcpServers": {
    "domainscan": {
      "command": "npx",
      "args": ["-y", "@domainscan/mcp-server"],
      "env": {
        "DOMAINSCAN_API_KEY": "your_api_key_here"
      }
    }
  }
}

The -y flag auto-confirms the npx install prompt so it runs non-interactively.

Troubleshooting

Tools not appearing in the AI client

• Restart the AI client after editing the config
• Check the config file JSON is valid (no trailing commas)
• Verify the API key is correct and active

“Tool call failed” errors

• Check your API key has not expired
• Verify you have network access to api.domainscan.in
• Try the npx command directly in terminal to see raw errors

Rate limit errors during multi-tool chains

• The MCP server handles rate limiting internally with backoff
• For very intensive use, contact DomainScan about higher-rate API access

For additional help, use Prism AI on the DomainScan website or contact support.