Skip to main content

Connect to CloudBase MCP

After connecting via the Model Context Protocol (MCP), you can operate CloudBase services directly in your AI chat—no need to switch between your IDE and the console.

CloudBase MCP supports Local Mode (run via npx on your machine) and Hosted Mode (connect to the cloud over HTTP). In Local Mode you can configure environment variables; in Hosted Mode you can control plugins via a URL.

See Connection Modes: Local vs Hosted.

Fastest start: Copy the AI Prompt below into your AI IDE.

Help me get set up with CloudBase. Do the following:
1. Open https://docs.cloudbase.net/skill.md and follow the instructions.
2. When done, tell me you're ready and suggest the most relevant next steps.

Prerequisites​

Before you start, make sure you have:

Node.js and a CloudBase environment ready

Node.js: Install Node.js v18.15.0 or later:

node --version

If you don’t have it installed, download it from the Node.js website.

CloudBase environment: Follow the doc Create a CloudBase environment. New users can start for free.

Connect Your AI Tool​

Pick your AI development tool and follow the setup guide:

Client

Configure your AI tool to connect with CloudBase capabilities. Supports local and hosted connection. See connection modes.

Step 1: Install / Configure CloudBase

Use project template (recommended) - Template includes MCP configuration and AI rulesView templates

Install in one click:

Add to Cursor

Or manual configuration:

Or add this configuration to .cursor/mcp.json:

json
1{
2  "mcpServers": {
3    "cloudbase": {
4      "command": "npx",
5      "args": ["@cloudbase/cloudbase-mcp@latest"],
6      "env": {
7        "INTEGRATION_IDE": "Cursor"
8      }
9    }
10  }
11}
Need help?View Cursor docs

Step 2: Chat with AI

Enter the following in your AI chat in order:

prompt
Read https://docs.cloudbase.net/skill.md and tell me what it can do
Use CloudBase Skills: 在 Web 应用中使用 CloudBase 文档数据库实现分页查询和复杂查询功能

Manually Confirm Tool Calls​

Most MCP clients (e.g. Cursor) will ask you to manually confirm every tool call. We recommend keeping this enabled and carefully reviewing the tool call details before execution.

Start Using It​

After setup, type in your AI chat:

Log in to CloudBase

Your AI will guide you through login and environment selection. Then you can start describing requirements in natural language and let the AI build your app.

🌏 International site users: if your environment is in the Tencent Cloud international site, declare the site first (TCB_SITE=intl); otherwise login goes through the China-site path and your international-site environment won't be visible. For all the ways to set it (MCP env, --site startup flag, specifying it in the conversation, CLI), see Connection Modes - International site users.

For the case where a China-site environment in the ap-singapore region is misdetected as international, see FAQ.

Examples​

End-to-end example:

Build a todo app with phone-number login using CloudBase, and deploy it

The AI can automatically: generate frontend and backend code, configure auth and the database, deploy cloud functions and static hosting, and return an access link.

Step-by-step examples:

  • "Build a WeChat Mini Program ordering system with CloudBase" — AI scaffolds pages, creates the cloud database, writes cloud functions, and guides preview
  • "Build a reservation management system that lets users place orders and admins update status" — AI models it with CloudBase PostgreSQL, configures permissions, and wires up frontend and backend
  • "Add phone-number login to a Web app" — AI enables the auth method and integrates the CloudBase SDK

Detailed requirement example:

Build a coffee reservation Mini Program with CloudBase that includes:
- Menu browsing and ordering
- Writing orders to the cloud database
- Cloud functions to handle order status
- Preview in WeChat DevTools

AI Skill Scenarios​

身份认证

身份认证：配置管理
配置管理认证提供商
身份认证：Web SDK
Web SDK 身份认证
身份认证：微信小程序
微信小程序身份认证
身份认证：App 集成
Android、iOS 原生应用
身份认证：Node.js 服务端
Node.js 服务端身份认证

数据库

文档数据库：Web SDK
Web SDK 文档数据库
文档数据库：小程序 SDK
小程序 SDK 文档数据库
关系型数据库：工具与规则
MySQL 关系型数据库
关系型数据库：Web SDK
Web SDK 关系型数据库
数据库：数据模型创建
复杂数据模型设计
数据库：App 集成
Android、iOS 原生应用
关系型数据库：PostgreSQL
PostgreSQL 开发与调试

后端开发

后端开发：云函数
Node.js 云函数
后端开发：云托管
Node.js、Python、Go、Java 等
后端开发：声明式部署
cloudbaserc 声明式部署
后端开发：Cloud API 操作
管控面 API 调用与 CAM 授权

应用集成

应用集成：Web 应用
React、Vue、Vite、浏览器验证
应用集成：最小 Web BaaS Demo
Lovable-like 最小前后端，Web SDK CRUD 优先
应用集成：微信小程序
小程序开发、tabBar、CloudBase
应用集成：微信生态集成
微信支付、公众号 OAuth
应用集成：UI 设计
专业界面设计指南
应用集成：App 集成
Android、iOS 原生应用
应用集成：云存储（Web SDK）
Web SDK 云存储

AI

在 Web 应用中集成 AI 功能
在 Web 应用中调用 AI
在 Node.js 后端中集成 AI 功能
在后端服务中调用 AI
在微信小程序中集成 AI 功能
在小程序中调用 AI
在应用中集成 AI Agent
Agent 服务构建与部署

开发工具

开发工具：Spec 工作流
需求分析、技术设计、任务拆分
开发工具：CloudBase 平台
平台知识库和最佳实践
开发工具：CloudBase CLI
tcb 命令行资源管理
开发工具：代码审查
项目落地检查与常见坑
开发工具：环境巡检与诊断
环境巡检、告警与故障诊断

Video Tutorials​

Want to learn CloudBase MCP via videos? See tutorials and practical cases.