MCP-Render
MCP server deployed on Render using SSE transport, providing tools for user management, ERP operations, SQLite database interaction, and server stats.
README
MCP Server - Render Deployment (SSE Transport)
This project is a sample Model Context Protocol (MCP) server that runs on Render using SSE (Server-Sent Events) transport.
Features
- HTTP + SSE Transport: Runs as a Render Web Service
- Tools:
get_users: List all usersget_user_by_id: Get a user by IDget_server_stats: Get server statscalculate: Basic math (add, subtract, multiply, divide)erp_list_customers: Mock ERP customers (read-only)erp_get_customer: Mock ERP customer by IDerp_list_orders: Mock ERP orders (read-only)erp_get_order: Mock ERP order by IDerp_list_invoices: Mock ERP invoices (read-only)erp_list_inventory: Mock ERP inventory (read-only)db_create: Create SQLite DB (optional schema)db_list_tables: List SQLite tables (read-only)db_table_schema: Show table schema (read-only)db_query: Run SELECT/WITH queries (read-only)db_exec: Run INSERT/UPDATE/DELETE/CREATE (write)
- Resources:
data://users: Users list (JSON)data://stats: Server stats (JSON)
Endpoints
| Endpoint | Method | Description |
|---|---|---|
/ |
GET | API info |
/health |
GET | Health check (for Render) |
/sse |
GET | SSE connection (MCP client connects here) |
/message |
POST | MCP messages (client to server) |
Local Development
# Install dependencies
npm install
# Build TypeScript
npm run build
# Start server
npm start
# Watch mode
npm run dev
The server runs on http://localhost:8080.
Deploy to Render
1. Push to GitHub
git init
git add .
git commit -m "MCP Server SSE"
git remote add origin https://github.com/yourusername/mcp-server-sse.git
git push -u origin main
2. Create a Render Web Service
- Log in to the Render Dashboard
- Select New > Web Service
- Connect your GitHub repo
- Use these settings:
- Runtime: Docker
- Plan: Free (or your preferred plan)
- Dockerfile Path:
./Dockerfile - Port:
8080
- Click Deploy
3. Deploy with Blueprint (Optional)
You can deploy using render.yaml:
- Go to Render Blueprints
- Select New Blueprint Instance
- Choose your repo
- Deploy
MCP Client Usage
To connect to the MCP server over SSE, the client must use the SSE URL.
Claude Desktop Config (example)
{
"mcpServers": {
"render-server": {
"url": "https://your-service.onrender.com/sse"
}
}
}
Local Development Config
{
"mcpServers": {
"local-server": {
"url": "http://localhost:8080/sse"
}
}
}
API Testing
Health Check
curl http://localhost:8080/health
SSE Connection (in a new terminal)
curl http://localhost:8080/sse
Send MCP Message
curl -X POST http://localhost:8080/message \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/list"
}'
Logging
Tool calls are logged as JSON lines. By default logs go to stdout.
Set LOG_FILE to write logs to a file instead.
SQLite
The server can create and query a local SQLite database.
Env vars:
SQLITE_FILE(default:./data/app.db)
Example usage:
db_createwithschema_sql: "CREATE TABLE items(id INTEGER PRIMARY KEY, name TEXT);"db_list_tablesdb_table_schemawithtable: "items"db_querywithsql: "SELECT * FROM items"db_execwithsql: "INSERT INTO items(name) VALUES('Elma'),('Armut')"
Project Structure
.
├── src/
│ └── index.ts # MCP server code (Express + SSE)
├── dist/ # Compiled JavaScript (after build)
├── Dockerfile # Docker image
├── render.yaml # Render blueprint
├── package.json # NPM dependencies
├── tsconfig.json # TypeScript config
└── README.md # This file
Technologies
- @modelcontextprotocol/sdk: MCP Server SDK
- Express: HTTP framework
- SSE: Server-Sent Events transport
- TypeScript: Type-safe development
- Docker: Containerization
- Render: Cloud deployment
License
MIT
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。
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。
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。
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。