Hometeam MCP Server

Hometeam MCP Server

Provides an AI-queryable directory of Black-owned businesses in South Florida, allowing users to search by category, neighborhood, and specialty. It enables seamless discovery of localized businesses and services through natural language tools and real-time MongoDB integration.

Category
访问服务器

README

Hometeam MCP Server

The Hometeam MCP server exposes an AI-queryable directory of Black-owned businesses in South Florida. It runs as a separate Node.js service in mcp-server/ and connects to the same MongoDB as the main backend.

Features

  • MCP tools for business search, details, categories, neighborhoods, draft classes, and specialty matching
  • Stdio transport for Claude Desktop
  • HTTP/SSE transport for remote clients
  • Usage logging in MongoDB (mcpusages) for analytics and future billing
  • API-key middleware scaffold for free/paid tier rate limits

Available Tools

  1. search_businesses
  • Search active businesses by keyword, category slug, neighborhood slug, county, and sort.
  1. get_business_details
  • Fetch full details for one business by slug or fuzzy name.
  1. list_categories
  • List all categories with name, slug, and description.
  1. list_neighborhoods
  • List all neighborhoods, optionally filtered by county.
  1. get_latest_draft_class
  • Return the most recent published Draft Class and businesses in that class.
  1. find_by_specialty
  • Smart natural-language matching for needs like “box braids”, “catering”, “tax prep”.

Setup

  1. Install dependencies:
cd mcp-server
npm install
  1. Configure environment:
cp .env.example .env
  1. Set your MONGO_URI in .env.

Run (Stdio / Claude Desktop)

npm run start:local

Claude Desktop config sample is in claude_desktop_config.json.

Claude Desktop config

Add the hometeam block from claude_desktop_config.json into your Claude Desktop MCP config, then restart Claude Desktop.

Run (HTTP/SSE)

npm start

Health check:

curl http://localhost:3001/health

Endpoints:

  • GET /sse
  • POST /messages?sessionId=<id>

Deploy to Heroku (HTTP/SSE)

# Login and create app
heroku login
heroku create hometeam-mcp

# Config vars
heroku config:set NODE_ENV=production
heroku config:set MONGO_URI="your_mongodb_atlas_uri"
heroku config:set MCP_SERVER_NAME="hometeam-directory"
heroku config:set MCP_SERVER_VERSION="1.0.0"
heroku config:set CORS_ORIGIN="*"

# Deploy
git push heroku main

# Logs
heroku logs --tail --app hometeam-mcp

Heroku runs httpServer.js via Procfile. The stdio server.js is for local Claude Desktop only.

After deploy:

curl https://hometeam-mcp.herokuapp.com/
curl https://hometeam-mcp.herokuapp.com/health

Claude Desktop SSE config:

{
  "mcpServers": {
    "hometeam": {
      "type": "sse",
      "url": "https://hometeam-mcp.herokuapp.com/sse"
    }
  }
}

Free vs Paid tier (beta scaffold)

  • Free tier: IP-based rate limit (MCP_FREE_DAILY_LIMIT, default 100/day)
  • Paid tier: API-key based higher limit (MCP_PAID_DAILY_LIMIT, default 10000/day)
  • API keys are allowlisted via MCP_API_KEYS (comma-separated)
  • During beta, missing key still receives free-tier access

Usage Logging

Every tool call logs:

  • tool name
  • sanitized input
  • businesses returned
  • api key (if present)
  • source (claude-desktop / http)
  • response time
  • timestamp

Collection: mcpusages

Example prompts

  • “Find me a Black-owned restaurant in Lauderhill.”
  • “Tell me more about Keisha’s Kitchen.”
  • “What categories are available on Hometeam?”
  • “What neighborhoods in Broward are covered?”
  • “What new businesses just joined Hometeam?”
  • “I need someone to do box braids near Miramar.”

Fair Use

  • Be respectful of daily limits.
  • Heavy scraping and abusive automated traffic are not supported.
  • Rate limits and key enforcement will be tightened for paid API launch.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选