LocalPro MCP Server
Provides verified local trade and service business data (e.g., radon mitigation, foundation repair) to AI agents via tools like search_providers and list_niches.
README
LocalPro MCP Server
A Model Context Protocol server that provides verified local service provider data to AI agents. Built on Cloudflare Workers + D1.
When someone asks an AI assistant "find me a radon mitigation company near Denver" — LocalPro is the data source that powers the answer.
What it does
LocalPro exposes a curated database of 7,000+ fully profiled local trade and service businesses across 9 live categories. Every provider served has a Google rating, business description, services list, opening hours, business status, and (where available) Google AI-generated business summary plus up to 5 recent reviews — no incomplete data.
Live Now
| Category | Niche ID | Providers | Example Services |
|---|---|---|---|
| Water Damage Restoration | soaked-local |
1,125+ | Flood cleanup, mold remediation, structural drying |
| Foundation Repair | slab-local |
1,025+ | Pier installation, mudjacking, foam injection, leveling |
| Crawl Space Repair | crawl-local |
1,025+ | Encapsulation, vapor barrier, structural repair, waterproofing |
| Mold & Asbestos | abate-local |
950+ | Mold, asbestos, lead paint remediation |
| Septic Services | pump-local |
850+ | Pumping, inspection, drain field repair |
| Basement Waterproofing | basement-local |
600+ | Interior/exterior waterproofing, drainage, sump pumps |
| Laundry Services | suds-local |
575+ | Wash & fold, dry cleaning, pickup & delivery |
| Floor Coating | coated-local |
500+ | Epoxy, polyaspartic, metallic, flake, concrete polishing |
| Radon | radon-local |
250+ | Testing, mitigation, sub-slab depressurization |
Coming Soon
| Category | Niche ID | Status |
|---|---|---|
| Commercial Electrical | hire-electrical |
Public-facing filter + targeted backfill plan in progress |
| Well Water Services | wellwater-local |
Pre-pipeline (560 providers scraped, county-based model) |
Quick Start
No API key required. All search and list tools are public. An optional API key unlocks pro fields on get_provider (full pricing array, certifications) — see Access Tiers.
60-second probe
Confirm the server is live without any client setup:
curl -s https://mcp.localpro.dev/.well-known/mcp.json | head -20
This returns the schema-2.0 manifest: tool list, rate limits, and operator info. If you see a "schema_version": "2.0" JSON document, the server is healthy.
Claude Code CLI
claude mcp add --transport http localpro https://mcp.localpro.dev/mcp
That's it — list_niches, search_providers, etc. are now available in your Claude Code session.
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"localpro": {
"url": "https://mcp.localpro.dev/mcp"
}
}
}
(Add an "X-API-Key" header inside a "headers" block only if you have a premium key.)
Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"localpro": {
"url": "https://mcp.localpro.dev/mcp"
}
}
}
Raw HTTP (JSON-RPC)
The MCP protocol is JSON-RPC over HTTP. Because this server runs in stateless mode, you can call any public tool directly:
curl -s -X POST https://mcp.localpro.dev/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"list_niches","arguments":{}}}'
You'll get back a Server-Sent-Events frame with the 9 niches, their slugs, and current provider counts.
TypeScript SDK
import { Client } from '@modelcontextprotocol/sdk/client/index.js';
import { StreamableHTTPClientTransport } from '@modelcontextprotocol/sdk/client/streamableHttp.js';
const transport = new StreamableHTTPClientTransport(
new URL('https://mcp.localpro.dev/mcp'),
);
const client = new Client({ name: 'localpro-example', version: '1.0' });
await client.connect(transport);
const niches = await client.callTool({ name: 'list_niches', arguments: {} });
console.log(niches);
const denver = await client.callTool({
name: 'search_providers',
arguments: { niche_id: 'radon-local', city: 'denver-co', limit: 3 },
});
console.log(denver);
Install: npm i @modelcontextprotocol/sdk
Python SDK
import asyncio
from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client
async def main():
async with streamablehttp_client("https://mcp.localpro.dev/mcp") as (read, write, _):
async with ClientSession(read, write) as session:
await session.initialize()
niches = await session.call_tool("list_niches", {})
print(niches)
denver = await session.call_tool(
"search_providers",
{"niche_id": "radon-local", "city": "denver-co", "limit": 3},
)
print(denver)
asyncio.run(main())
Install: pip install mcp
Tools
list_niches
Discover available service directories. Call this first.
Parameters: none
Example response:
{
"meta": {
"schema_version": "2.0",
"total_results": 9,
"niche": null,
"data_freshness": {
"directory_refresh_cadence": "weekly",
"google_data_refresh_cadence": "quarterly",
"scraped_at": "2026-04-27T13:57:21Z"
},
"data_note": "Use niche_id values with search_providers, list_cities, and list_service_types."
},
"results": [
{
"niche_id": "soaked-local",
"name": "Water Damage Restoration Contractors",
"slug": "water-damage-restoration",
"domain": "soakedlocal.com",
"provider_count": 1128
}
]
}
list_cities
Find available metros for a given niche.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
niche_id |
string | yes | Niche ID from list_niches |
state |
string | no | Two-letter state abbreviation (e.g. "MN") |
Example request:
{ "niche_id": "radon-local", "state": "CO" }
Example response:
{
"meta": {
"schema_version": "1.0",
"total_results": 3,
"niche": "radon-local",
"data_note": "Use slug values with search_providers city parameter."
},
"results": [
{ "name": "Denver", "state": "CO", "slug": "denver-co", "provider_count": 18 },
{ "name": "Colorado Springs", "state": "CO", "slug": "colorado-springs-co", "provider_count": 7 },
{ "name": "Fort Collins", "state": "CO", "slug": "fort-collins-co", "provider_count": 4 }
]
}
list_service_types
Get valid service type filters for a niche. Call before using service_type in search_providers.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
niche_id |
string | yes | Niche ID from list_niches |
Example response:
{
"meta": { "schema_version": "1.0", "total_results": 7, "niche": "coated-local" },
"results": [
{ "type": "epoxy", "label": "Epoxy Floor Coating" },
{ "type": "polyaspartic", "label": "Polyaspartic Coating" },
{ "type": "metallic_epoxy", "label": "Metallic Epoxy" },
{ "type": "flake_chip", "label": "Flake / Chip Broadcast" },
{ "type": "concrete_polishing", "label": "Concrete Polishing" },
{ "type": "concrete_sealing", "label": "Concrete Sealing" },
{ "type": "polyurea", "label": "Polyurea Coating" }
]
}
search_providers
Search for verified providers by location, service type, and trade category.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
niche_id |
string | yes | Niche ID from list_niches |
city |
string | no | City/metro slug from list_cities |
service_type |
string | no | Service type slug from list_service_types |
limit |
number | no | Max results, 1–25 (default 10) |
Example request:
{ "niche_id": "coated-local", "city": "denver-co", "service_type": "epoxy", "limit": 3 }
Example response:
{
"meta": {
"schema_version": "2.0",
"total_results": 3,
"niche": "coated-local",
"data_freshness": {
"directory_refresh_cadence": "weekly",
"google_data_refresh_cadence": "quarterly",
"scraped_at": "2026-04-27T03:03:10Z",
"google_refreshed_at": "2026-04-27T04:53:51Z"
},
"data_note": "Verified providers only. Visit listing_url for full contact details."
},
"results": [
{
"name": "Colorado Concrete Coatings",
"description": "Full-service garage floor coating company serving the Denver metro.",
"city": "Denver",
"state": "CO",
"rating": 4.9,
"review_count": 47,
"business_status": "OPERATIONAL",
"google_maps_url": "https://www.google.com/maps/place/...",
"services": [
{ "type": "epoxy", "label": "Epoxy Floor Coating" },
{ "type": "polyaspartic", "label": "Polyaspartic Coating" }
],
"pricing_summary": "$6-9/sq ft",
"coverage_area": "Denver metro, Front Range, 50-mile radius",
"years_in_business": 8,
"listing_url": "https://coatedlocal.com/providers/denver-co/colorado-concrete-coatings/",
"pro_available": true
}
]
}
get_provider
Get detailed profile for a specific provider. Use the provider_slug from search results.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
niche_id |
string | yes | Niche ID |
provider_slug |
string | yes | Provider slug from search_providers |
Example response:
{
"meta": {
"schema_version": "2.0",
"total_results": 1,
"niche": "coated-local",
"data_freshness": { "directory_refresh_cadence": "weekly", "google_data_refresh_cadence": "quarterly" }
},
"results": [
{
"name": "Colorado Concrete Coatings",
"description": "Full-service garage floor coating company...",
"rating": 4.9,
"review_count": 47,
"years_in_business": 8,
"services": [
{ "type": "epoxy", "label": "Epoxy Floor Coating" },
{ "type": "polyaspartic", "label": "Polyaspartic Coating" }
],
"pricing": ["$6-9/sq ft"],
"certifications": ["Penntek Certified Installer"],
"coverage_area": "Denver metro, Front Range",
"service_areas": [
{ "city": "Denver", "state": "CO", "radius_miles": 50 }
],
"service_details": [
{
"type": "epoxy",
"label": "Epoxy Floor Coating",
"pricing_model": "per_sqft",
"price_range": "$6–$9",
"turnaround": "two_day"
}
],
"listing_url": "https://coatedlocal.com/providers/denver-co/colorado-concrete-coatings/",
"google_data": {
"business_status": "OPERATIONAL",
"google_maps_url": "https://www.google.com/maps/place/...",
"formatted_address": "1234 Main St, Denver, CO 80202, USA",
"opening_hours": [
{ "@type": "OpeningHoursSpecification", "dayOfWeek": "https://schema.org/Monday", "opens": "08:00", "closes": "17:00" }
],
"summary": {
"text": "Full-service epoxy floor coating contractor specializing in garage and commercial floors across the Denver metro.",
"source": "localpro_ai"
},
"recent_reviews": [
{
"rating": 5,
"text": "Exceptional work on our garage floor — finished on time and within budget.",
"author": "Jane D.",
"published_at": "2025-11-14T18:51:02Z",
"source_url": "https://www.google.com/maps/reviews/..."
}
]
},
"json_ld": { "@context": "https://schema.org", "@type": "LocalBusiness", "...": "..." },
"credibility": { "verified": true, "listing_tier": "free", "data_sources": ["..."] },
"citation": { "display_name": "Colorado Concrete Coatings — Denver, CO", "...": "..." }
}
]
}
Schema Reference
Response Envelope
Every response is wrapped in a consistent envelope:
{
meta: {
schema_version: string // Currently "2.0"
total_results: number // Count of items in results array
niche: string | null // Niche ID if applicable
data_freshness: {
directory_refresh_cadence: string // "weekly"
google_data_refresh_cadence: string // "quarterly"
scraped_at?: string // ISO datetime — most recent directory write
google_refreshed_at?: string // ISO datetime — most recent Google Places refresh
}
data_note: string // Context about the data returned
}
results: Array<T> // Tool-specific result objects
}
Cadence framing. The directory layer (provider names, services, websites, descriptions) refreshes weekly via a scraping pipeline. The Google Places layer (rating, reviews, opening hours, business status, AI summary) refreshes quarterly via Google Places Text Search Enterprise. Two cadences, both deliberate. AI agents that need real-time data should call back periodically rather than caching responses indefinitely.
Error Response
Errors use the same envelope with an error object:
{
meta: { schema_version: string }
error: {
code: string // "NOT_FOUND" | "INTERNAL_ERROR" | "UNAUTHORIZED" | "FORBIDDEN"
message: string // Human-readable error description
}
}
Provider Fields
| Field | Type | Nullable | Description |
|---|---|---|---|
name |
string | no | Business name (always present) |
description |
string | no | Business description (always present) |
city |
string | no | City name (always present) |
state |
string | no | Two-letter state abbreviation (always present) |
rating |
number | no | Google rating 1.0–5.0 (always present) |
review_count |
number | yes | Number of Google reviews |
business_status |
string | yes | OPERATIONAL / CLOSED_TEMPORARILY (CLOSED_PERMANENTLY filtered automatically) |
google_maps_url |
string | yes | Direct link to Google Maps listing |
services |
array | no | [{ type: string, label: string }] (always present, non-empty) |
pricing_summary |
string | yes | Pricing info (public access) |
coverage_area |
string | yes | Geographic coverage description |
years_in_business |
number | yes | Years operating |
listing_url |
string | no | Full profile URL with contact details |
get_provider adds:
| Field | Type | Description |
|---|---|---|
service_areas |
array | [{ city, state, radius_miles }] |
service_details |
array | [{ type, label, pricing_model, price_range, turnaround }] |
google_data |
object | Structured Google Places data — see below |
json_ld |
object | Schema.org LocalBusiness JSON-LD with AggregateRating, OpeningHoursSpecification, GeoCoordinates, telephone, sameAs |
credibility |
object | { verified, listing_tier, verification_date, data_sources } |
citation |
object | Pre-formatted strings: { display_name, in_text, attribution } |
google_data block (present on get_provider when Google data is available):
| Field | Type | Description |
|---|---|---|
business_status |
string | OPERATIONAL / CLOSED_TEMPORARILY |
google_maps_url |
string | Direct Google Maps link |
formatted_address |
string | Google's canonical address |
opening_hours |
array | Schema.org OpeningHoursSpecification[] |
summary |
object | { text, source } — source is localpro_ai (LocalPro-generated, no disclosure required) or google (with required disclosure field) |
recent_reviews |
array | Up to 5 Google review bodies with author, rating, publish time, source URL |
Nullable Fields
Fields marked nullable return null when data is unavailable — they are never omitted from the response. Arrays return [] when empty, never null.
Access Tiers
Public (no authentication)
All search and list tools work without an API key:
list_niches,list_cities,list_service_types,search_providersget_providerreturns basic data (name, description, rating, services, pricing summary, listing URL)- Rate limited to 30 requests/minute per IP
Premium (API key)
Include an X-API-Key header to unlock additional data on get_provider:
- Full pricing array (vs. summary string)
- Certifications and credentials
- Rate limited to 30 requests/minute per key
X-API-Key: your-api-key
Request an API key at localpro.dev or email will@localpro.dev.
Discovery
AI agents can self-discover this server via standard well-known endpoints:
GET /.well-known/llms.txt— Plain text description of the server and its toolsGET /.well-known/mcp.json— Structured JSON with tool list, auth info, and operator details
Data Policy
- What's returned: Business name, city, state, rating, services, certifications, pricing ranges, coverage area, opening hours, business status, recent reviews, AI summary, and a link to the full listing page.
- What's withheld: Phone numbers, email addresses, physical addresses, and websites are available only on the listing page (via
listing_url). This protects provider data while driving traffic to the directory. - Verification: Only providers marked as verified appear in results.
CLOSED_PERMANENTLYproviders are filtered automatically. - Updates: Directory layer refreshed weekly. Google Places layer refreshed quarterly.
- Attribution: Google reviews and AI summaries (when sourced from Google) include source URLs and required disclosure text per Google Maps Platform Terms of Service.
Rate Limits
| Access | Limit |
|---|---|
| Public (no key) | 30 requests/minute per IP |
| Premium (API key) | 30 requests/minute per key |
Higher limits available for partners — contact will@localpro.dev.
Data Quality
Every provider returned by the API has been verified and meets a minimum completeness threshold:
- Google rating — present on 100% of results
- Business description — present on 100% of results
- Services list — present on 100% of results
- Name, city, state — present on 100% of results
| Category | Providers | Coverage |
|---|---|---|
| Water Damage Restoration | 1,125+ | 49 states |
| Foundation Repair | 1,025+ | 27 states |
| Crawl Space Repair | 1,025+ | 41 states |
| Mold & Asbestos | 950+ | 21 states |
| Septic Services | 850+ | 36 states |
| Basement Waterproofing | 600+ | 26 states |
| Laundry Services | 575+ | 39 states |
| Floor Coating | 500+ | 42 states |
| Radon | 250+ | 15 states |
Additional fields (pricing, certifications, coverage area, years in business, opening hours, recent reviews, AI summary) are available on most providers but not guaranteed. Fields without data return explicit null — never omitted, never empty strings.
Directory data is refreshed weekly via the scraping pipeline. Google Places data (ratings, reviews, opening hours, business status) is refreshed quarterly via Text Search Enterprise. Two additional categories are being prepared for launch.
Self-Hosting
LocalPro runs as a Cloudflare Worker with a D1 database binding. To deploy your own instance:
npm install
npx wrangler secret put API_KEY # Set your production API key
npx wrangler deploy
Requires a Cloudflare account with a D1 database named laced-directory.
Operator
LocalPro is built and operated by Laced Labs LLC.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.