tencentcloudbase/cloudbase-mcp
Overview
This skill provides essential CloudBase (Tencent CloudBase) development guidelines and workflow rules for building web apps, WeChat mini programs, backend services, databases, storage, deployment, and AI features. It emphasizes using CloudBase MCP (Model Context Protocol) or mcporter to inspect and run CloudBase tools, and it guides developers on platform-specific constraints and recommended skill mappings. Read this before starting or contributing to any CloudBase project.
How this skill works
The skill inspects the project context and maps user requests to CloudBase capabilities (hosting, functions, databases, auth, storage, AI). It enforces a workflow: understand tools first via MCP (or mcporter), authenticate with device-code login, then run tool descriptions before calling actions. It recommends the right sub-skills for SDK integration, deployment, and platform-specific patterns.
When to use it
- Starting a new CloudBase full-stack web app or migrating an existing app to CloudBase
- Building or deploying WeChat mini programs that rely on wx.cloud and OPENID
- Creating or deploying cloud functions, container services (CloudRun), or static hosting
- Implementing database models (NoSQL collections or MySQL tables) and security rules
- Adding authentication providers or using AI models (text/chat/image) with streaming
- Working in environments where MCP is not available and mcporter CLI is required
Best practices
- Always run npx mcporter describe cloudbase (or MCP describe) before managing or deploying CloudBase
- Prefer device-code login via MCP auth tool; do not hard-code SecretId/SecretKey/EnvId in configs
- Install CloudBase MCP in your agent/workspace for consistent environment management and deployments
- Deploy backend components before frontend previews when backend services are required
- Use SDKs directly for database access where possible; use cloud functions only when necessary
- Map user needs to specific CloudBase skills (auth-web, no-sql-web-sdk, ai-model-web, etc.)
Example use cases
- Deploy a React site to CloudBase static hosting and wire in Web SDK authentication
- Create a WeChat mini program using wx.cloud with cloud functions and document DB
- Set up a Node.js cloud function that calls a CloudBase AI model with streaming output
- Provision a MySQL database, design tables, and use relational-database-tool for management
- Use mcporter to start device-code login, bind an envId, then deploy cloud functions
FAQ
No. Use MCP device-code auth or interactive login via mcporter to avoid embedding secrets.
What if my environment doesn't support MCP?
Use the mcporter CLI to call CloudBase MCP tools (npx mcporter call cloudbase.<tool> key=value).
12 skills
This skill guides CloudBase MCP usage for efficient full-stack development, including authentication, databases, storage, and serverless deployment across web,
This skill helps you create, rewrite, and evaluate SKILL.md files with clear triggers, structure, and testable behavior for reliable agent skills.
This skill helps you build and deploy AI agents using the CloudBase AG-UI protocol in TypeScript, including adapters, servers, and web UI.
This skill helps you understand CloudBase platform concepts, configure authentication, deploy functions, manage databases, and optimize data models across web
This skill helps you query, create, update, and delete CloudBase documents in WeChat MiniProgram apps with efficient, scalable database operations.
This skill guides structured software development from requirements to task execution using EARS syntax and UI-design references to ensure quality.
This skill helps you develop CloudBase Run backends with long connections and AI agents, supporting multi-language runtimes and secure, scalable deployments.
This skill helps you manage CloudBase cloud storage in web apps by uploading, generating temporary URLs, downloading, and deleting files with the Web SDK.
This skill helps you develop, deploy, and manage CloudBase cloud functions by guiding runtime selection, deployment, logging, and HTTP access configuration.
This skill enables browser apps to generate and stream AI text using CloudBase JS SDK, with Hunyuan or DeepSeek models.
This skill enables browser apps to initialize CloudBase Relational Database with a single shared db client and use Supabase-style queries.
This skill guides calling AI models with CloudBase across web, Node, and WeChat Mini Program, enabling text, streaming, and image workflows.
