sthan-io/mcp-server

npm version License: MIT

MCP server for sthan.io — give your AI assistant the ability to verify, parse, autocomplete, and geocode US addresses, and look up IP geolocation. Works with Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, and any MCP-compatible client.

8 tools · TypeScript · stdio transport · Free tier, no credit card required.

Why use it

• Real US postal data — addresses are verified against authoritative postal records, not heuristics. Returns delivery point validation (DPV), ZIP+4, residential/commercial flag.
• Sub-100ms autocomplete — typeahead-grade response times for address, city, and ZIP suggestions.
• Strong geocoding — forward geocoding returns accuracy classification (rooftop / interpolated / centroid / approximate) plus a confidence score, so your AI knows when to trust the result.
• IPv4 + IPv6 geolocation — country, region, city, coordinates, timezone, postal code.
• Free tier with no credit card — get a key in under a minute at sthan.io/dashboard.

Tools

Quick start

1. Get a free API key

Sign up at sthan.io and create a key from the dashboard. No credit card.

2. Add to your MCP client

Claude Desktop / Claude Code (~/.claude/mcp.json or project .claude/mcp.json):

{
  "mcpServers": {
    "sthan": {
      "command": "npx",
      "args": ["-y", "@sthan/mcp-server"],
      "env": { "STHAN_API_KEY": "sthan_test_your_key_here" }
    }
  }
}

Cursor (.cursor/mcp.json):

{
  "mcpServers": {
    "sthan": {
      "command": "npx",
      "args": ["-y", "@sthan/mcp-server"],
      "env": { "STHAN_API_KEY": "sthan_test_your_key_here" }
    }
  }
}

VS Code (settings.json):

{
  "mcp.servers": {
    "sthan": {
      "command": "npx",
      "args": ["-y", "@sthan/mcp-server"],
      "env": { "STHAN_API_KEY": "sthan_test_your_key_here" }
    }
  }
}

Example prompts

Once configured, just ask your AI assistant naturally:

• "Is 123 Main St, New York, NY 10001 a real, deliverable address?"
• "Parse this address: apt 2b 500 broadway new york ny"
• "What are the coordinates of 1600 Pennsylvania Ave, Washington DC?"
• "What's the nearest address to 40.7128, -74.0060?"
• "Where is IP 8.8.8.8 located?"
• "Suggest US cities starting with 'San Fr'."

Environment variables

Packages

This monorepo publishes two packages:

Development

git clone https://github.com/sthan-io/mcp-server.git
cd mcp-server
npm install
npm run build --workspaces

Then run the server with STHAN_API_KEY=sthan_test_... node packages/mcp-server/dist/index.js.

Links

• Homepage: https://sthan.io
• API docs: https://sthan.io/docs
• OpenAPI spec: https://api.sthan.io/openapi.json
• AI reference (llms-full.txt): https://api.sthan.io/llms-full.txt
• Pricing: https://sthan.io/pricing/united-states
• Support: https://sthan.io/support
• Glama: https://glama.ai/mcp/servers/sthan-io/mcp-server

License

MIT — see LICENSE.