Skip to main content

CloudBase CLI Configuration Guide

CloudBase CLI is a unified command-line tool that integrates multiple mainstream AI programming tools, supports project-level configuration, and includes built-in CloudBase MCP and AI development rules.

Core Advantages:

  • Unified Management: One command manages multiple AI programming CLI tools, no need to switch between multiple tools
  • Built-in MCP and Rules: Ready to use out of the box, no need to manually configure MCP server and AI rules
  • Project-level Configuration: Supports project-level configuration for easier team collaboration
  • Multi-model Support: Supports built-in and custom various large models, including Kimi K2, Zhipu GLM-4.5, etc.

For example:

  • "Create user table" - AI automatically creates database collection
  • "Deploy this function" - AI automatically uploads and configures cloud function
  • "Deploy frontend to CDN" - AI automatically uploads files and configures hosting

No need to switch to cloud console, all operations can be completed in command line using natural language.

Prerequisites​

Before starting configuration, please ensure the following conditions are met:

Node.js environment and CloudBase environment are ready

Node.js Environment: Ensure Node.js v18.15.0 or higher is installed:

node --version

If not installed, please download and install from Node.js official website.

CloudBase Environment: Please refer to the documentation Create CloudBase Environment. New users can sign up for free.

Install CloudBase CLI​

Install globally using npm/yarn/pnpm:

npm i -g @cloudbase/cli

Or using Yarn 1.x:

yarn global add @cloudbase/cli

Or Pnpm:

pnpm add -g @cloudbase/cli

If there are issues accessing the official npm registry, you can use the domestic mirror:

npm i -g @cloudbase/cli --registry=http://mirrors.cloud.tencent.com/npm/

Initialize AI Configuration​

Run CloudBase AI CLI in terminal:

# Start configuration wizard
tcb ai

On first launch, the configuration wizard will guide you through:

  • Select AI tool (CodeBuddy Code, Claude Code, OpenAI Codex, aider, Qwen Code)
  • Complete AI tool configuration (supports custom LLM or using LLM in CloudBase environment)

After configuration is complete, you can start using AI tools for assisted development. You can run tcb ai --setup later to switch tools and models.

Client

Configure your MCP client to connect with your CloudBase environment

Step 1: Installation

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

安装 CloudBase CLI：

bash
npm i -g @cloudbase/cli

初始化配置：

bash
tcb ai

配置向导会引导你完成 AI 工具的配置。CloudBase CLI 内置了 MCP 和 AI 开发规则，无需手动配置。

开始使用：

bash
tcb ai
Need help?View CloudBase CLI docs

Step 2: Verify connection

After configuration, enter the following in your AI chat to verify:

prompt
检查 CloudBase 工具是否可用, 下载 CloudBase AI 开发规则到当前项目

Get Started​

# Start AI tool
tcb ai

# Or preset prompt
tcb ai -- "Help me create an order management system"

# Or use non-interactive mode
tcb ai -- -p "Hello"

# View all available commands
tcb ai --help

Supported AI Tools​

CloudBase CLI supports the following AI programming tools:

AI ToolCommand Parameter
CodeBuddy Code-a codebuddy
Claude Code-a claude
OpenAI Codex-a codex
aider-a aider
Qwen Code-a qwen

Project-level Configuration​

CloudBase CLI supports project-level configuration, configuration files are automatically saved in the project root directory. This allows team members to share the same configuration and improve collaboration efficiency.

Common Questions​

Q: How to switch AI tools? A: Run tcb ai --setup to rerun the configuration wizard and select different AI tools.

Q: How to view current configuration? A: Run tcb ai --config to view current configuration information.

Q: How to reset configuration? A: Run tcb ai --reset to reset all configuration.

More questions please check: Complete FAQ

相关资源