LocationLists MCP server

No connector needed: any AI assistant can answer a location question by opening https://locationlists.com/find?q= — counts, a preview, a price and a card checkout link. Connect the MCP for repeat use or wallet payments.

Answer from a link (no connector)

Open the link. No account, no key, no setup. The answer comes back with how many locations match, a few real rows, the price and a checkout link.

• 100 retail stores within 25 miles of Los Angeles: https://locationlists.com/find?category=retail&near=Los+Angeles+CA&radius=25&total=100
• Dental practices in Richmond, VA: https://locationlists.com/find?dataset=dental-practice-list&city=Richmond&state=VA
• STIHL dealers within a 30-minute drive of zip 23219: https://locationlists.com/find?dataset=stihl-dealers&near=23219&drive=30

Or set no parameters at all and put the question itself in q=, in plain words.

Everything below is for repeat use and wallet payments.

Listed on Smithery · Glama · Official MCP Registry · x402scan

Business location data for AI agents. Search, sample, count and buy ready-to-use lists of US business locations: 966 datasets compiled from each brand's official store locator or the public register that publishes the data.

Filter by city, state, zip, a radius around any place, or any column in the file. Buy a full CSV by card, or let an agent pay per row or buy the whole file in USDC on Base via x402, with no account or API key. A failed or empty query costs nothing.

This repository holds the registry listing (server.json) and client setup notes for the hosted LocationLists MCP server. The server runs at locationlists.com; there is nothing to install or run locally.

What is in the catalog

Figures are read from the live catalog.json (generated 2026-09-16):

• 966 datasets plus 2 bundles, 15,184,612 US business locations in total (overlapping slices counted once)
• Categories: Breakfast, Equipment, Financial, Furniture, Government, Grills, Hardware, Healthcare, Industrial, Mattresses, Nonprofits, Other Transactions, Outdoor Furniture, Retail
• Each dataset is compiled from the official locator, association directory or public register that publishes it
• One-time prices from $9 to $1,299 per dataset, by record count

Tools

All nine tools are on /mcp. Tools carry readOnlyHint / destructiveHint / openWorldHint annotations. initialize returns serverInfo.name = "LocationLists".

Free tools:

Card checkout (Stripe):

Agent payments (x402, USDC on Base):

Agents can pay with x402

query_locations and buy_dataset use x402, the HTTP 402 payment protocol. An agent that holds USDC on Base can buy data with no account, API key or checkout page:

1. Call the tool without payment. The server answers with the exact price and x402 payment requirements (USDC on Base mainnet, settled through the Coinbase CDP facilitator).
2. Sign the payment with the agent's wallet and call the tool again with the payment attached.
3. The server returns the rows (query_locations) or the permanent download link (buy_dataset). Per-row queries settle only after the rows are produced, so a failed query costs nothing.

Pricing rules the agent can rely on:

• count_locations is free and reports how many rows a filter matches and what they would cost, before any payment.
• Per-row prices are set per dataset at roughly twice the list price spread over its record count, so a small slice of a large file costs cents. You pay for the limit you request.
• Past a few hundred rows, buy_dataset is cheaper than assembling the file row by row, and it is complete.

Clients without a wallet can use get_quote → create_checkout → check_order and pay by card instead.

Client setup

Claude Code

claude mcp add --transport http locationlists https://locationlists.com/mcp

Cursor (~/.cursor/mcp.json)

{
  "mcpServers": {
    "locationlists": {
      "url": "https://locationlists.com/mcp"
    }
  }
}

VS Code (.vscode/mcp.json)

{
  "servers": {
    "locationlists": {
      "type": "http",
      "url": "https://locationlists.com/mcp"
    }
  }
}

Claude.ai and ChatGPT

• Claude.ai: Settings → Connectors → Add custom connector → URL https://locationlists.com/mcp, no OAuth.
• ChatGPT: Settings → Connectors (developer mode) → add https://locationlists.com/mcp/chatgpt, no auth. The read-only profile is used there because in-chat purchase of digital goods is not permitted in ChatGPT.

Any MCP client, by hand

curl -s https://locationlists.com/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_datasets","arguments":{"query":"bobcat"}}}'

A quick path through the tools: search_datasets {query:"bobcat"} → get_sample {slug:"bobcat-dealers"} → count_locations {dataset:"bobcat-dealers", state:"TX"} → get_quote {slugs:["bobcat-dealers"]} → create_checkout {slug:"bobcat-dealers"} → check_order {sessionId}.

Plain HTTP endpoints (no MCP client needed)

Example: bobcat-dealers/dataset.json, bobcat-dealers/sample.csv.

Links

• Site: https://locationlists.com
• Privacy policy: https://locationlists.com/privacy
• Data license and terms: https://locationlists.com/license
• Glama: https://glama.ai/mcp/connectors/io.github.kylehawke-stack/locationlists
• Contact: kyle@locationlists.com

License

The files in this repository (README, server.json) are MIT licensed; see LICENSE. The datasets sold through the server are not covered by that license; they are governed by https://locationlists.com/license.