Lead Stampede MCP Server
Exposes SMB business data as tools for AI agents, enabling retrieval of business profiles, services, availability, and reviews.
README
Lead Stampede MCP Server
The live backend that powers Lead Stampede Agent Cards. Exposes SMB business data as tools that AI agents can call over HTTP.
What this is
An Express server that implements four MCP tools:
| Tool | What it returns |
|---|---|
get_business_profile |
Name, description, industry, service area, contact info |
get_services |
List of services + pricing summary |
get_availability |
Hours + booking URL (or phone/email fallback) |
get_reviews |
Review count, average rating, summary |
Each request is authenticated via an X-Agency-API-Key header that maps to a
row in the agencies table in Supabase. Clients are scoped to their agency,
so one agency can never read another's data.
Setup
1. Install
npm install
2. Configure environment
cp .env.example .env
Then edit .env and fill in:
SUPABASE_URL— your project URL (e.g.https://xlhxnlhxeprtzwbyepza.supabase.co)SUPABASE_SERVICE_ROLE_KEY— from Supabase → Project Settings → API →service_rolesecretPORT— defaults to 3000
Never commit .env to git. It's already in .gitignore.
3. Run locally
npm run dev
You should see:
Lead Stampede MCP server listening on port 3000
Testing it works
Health check (no auth)
curl http://localhost:3000/health
# → {"status":"ok","timestamp":"..."}
List available tools (auth required)
curl -H "X-Agency-API-Key: YOUR_AGENCY_API_KEY" \
http://localhost:3000/mcp/tools
Call a tool
curl -X POST \
-H "X-Agency-API-Key: YOUR_AGENCY_API_KEY" \
-H "Content-Type: application/json" \
-d '{"arguments":{"client_slug":"grandinetti-molinar-law"}}' \
http://localhost:3000/mcp/tools/get_business_profile
Expected response:
{
"tool": "get_business_profile",
"result": {
"business_name": "Grandinetti & Molinar Law",
"description": "Austin-based law firm founded in 2002...",
"industry": "legal",
...
},
"response_ms": 42
}
Deploying to Railway
- Push this folder to a new GitHub repo.
- In Railway: New Project → Deploy from GitHub repo.
- Select this repo.
- Under Variables, add:
SUPABASE_URLSUPABASE_SERVICE_ROLE_KEY
- Railway auto-detects Node.js, installs deps, and runs
npm start. - Once deployed, Railway gives you a public URL like
mcp-production.up.railway.app. - Point a custom subdomain (e.g.
mcp.leadstampede.io) at it via Railway → Settings → Domains.
Architecture
AI agent → POST /mcp/tools/:name → auth.js → tools.js → supabase.js
↓
analytics.js (fire-and-forget)
↓
mcp_tool_calls table
Next steps
- Build the Cloudflare Worker that serves Agent Cards at
agentcards.leadstampede.io/{slug}. - Build the onboarding form that populates
clientsrows in Supabase. - Build the agency portal dashboard.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。