Skip to main content

CloudRun Development & Deployment

CloudBase AI ToolKit includes built-in support for CloudRun development and deployment. CloudRun lets you deploy and run various backend services easily, supporting long-lived connections, file uploads, and multiple languages.

This plugin is enabled by default—you just need to describe what you want to do in natural language.

New: AI Agent Development​

You can now develop AI agents based on function-mode CloudRun, making it easy to create and deploy personalized AI applications.

When should you use CloudRun?​

Use CloudRun when you need:

  • Real-time communication: WebSocket, SSE, streaming responses
  • Long-running tasks: background processing
  • Multiple languages: Java, Go, PHP, Python, Node.js, etc.
  • AI agents: personalized AI application development

Which mode should you choose?​

Function mode: recommended for beginners. Supports Node.js, has built-in WebSocket support, supports local debugging, fixed port 3000.

Container mode: suitable for existing projects. Supports any language, requires a Dockerfile.

Quick Start​

1) List available templates​

List available CloudRun templates

2) Create a new project​

Create a project named my-service from the helloworld template

3) Run locally (function mode)​

Run my-service locally on port 3000

4) Deploy to the cloud​

Deploy my-service with public access, CPU 0.5 core, memory 1GB

5) Create an AI agent​

Create an agent named my-agent for customer support chat

Common Scenarios​

Mini Program backend​

Create a function-mode service with WebSocket support for Mini Program chat

Java Spring Boot app​

Deploy a Spring Boot app that provides REST APIs

Go microservice​

Create a high-performance Go microservice to handle user authentication

Python data processing​

Deploy a Python service to process data on a schedule and generate reports

PHP Laravel app​

Deploy a Laravel app and expose a full admin backend

AI agent app​

Create an agent to handle user inquiries and provide personalized services

Access your service​

After deployment, you can access your service in these ways:

Call from WeChat Mini Program (recommended):

const res = await wx.cloud.callContainer({
  config: { env: "your-env-id" },
  path: "/api/data",
  method: "POST",
  header: { "X-WX-SERVICE": "my-service" }
});

Call from Web apps:

import cloudbase from "@cloudbase/js-sdk";

const app = cloudbase.init({ env: "your-env-id" });

const res = await app.callContainer({
  name: "my-service",
  method: "GET",
  path: "/health"
});

Direct HTTP access:

curl https://your-service-domain.com

AI Agent Development​

Create an agent​

Create an agent named customer-service for customer support chat

Run an agent locally​

Run the customer-service agent locally on port 3000

Call an agent​

// Call from a Web app
const app = cloudbase.init({ env: "your-env-id" });
const ai = app.ai();

const res = await ai.bot.sendMessage({
  botId: "ibot-customer-service-demo",
  msg: "Hello, I need help"
});

for await (let x of res.textStream) {
  console.log(x);
}
# CLI test
curl 'http://127.0.0.1:3000/v1/aibot/bots/ibot-customer-service-demo/send-message' \
  -H 'Accept: text/event-stream' \
  -H 'Content-Type: application/json' \
  --data-raw '{"msg":"Hello"}'