All skills
microsoft avatar

/azure-prepare

@7995851 official

Prepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. USE ONLY when the user explicitly wants to use azd as the deployment tool, or the project already has an azure.yaml file. DO NOT USE FOR: non-azd deployments, Python App Service code-only deploys (use python-appservice-deploy), or cross-cloud migration (use azure-cloud-migrate). WHEN: prepare app for azd, create azure.yaml, set up azd infrastructure, modernize app for Azure with azd, deploy with azd, function app, timer trigger, service bus trigger, event-driven function, managed identity, generate Bicep, generate Terraform, create and deploy to Azure.

Use this Skill: https://skilld.dev/gh/microsoft/github-copilot-for-azure/azure-prepare

This session only. Nothing lands on disk.

referencesservicesdurable-task-schedulerjavascript.md

≈1.4k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Durable Task Scheduler — JavaScript

Learn More

Durable Functions Setup

Required npm Packages

{
  "dependencies": {
    "@azure/functions": "^4.0.0",
    "durable-functions": "^3.0.0"
  }
}

💡 Finding latest versions: Run npm view durable-functions version or check npmjs.com/package/durable-functions for the latest stable release.

host.json

{
  "version": "2.0",
  "extensions": {
    "durableTask": {
      "hubName": "default",
      "storageProvider": {
        "type": "durabletask-scheduler",
        "connectionStringName": "DURABLE_TASK_SCHEDULER_CONNECTION_STRING"
      }
    }
  },
  "extensionBundle": {
    "id": "Microsoft.Azure.Functions.ExtensionBundle",
    "version": "[4.*, 5.0.0)"
  }
}

local.settings.json

{
  "IsEncrypted": false,
  "Values": {
    "FUNCTIONS_WORKER_RUNTIME": "node",
    "AzureWebJobsStorage": "UseDevelopmentStorage=true",
    "DURABLE_TASK_SCHEDULER_CONNECTION_STRING": "Endpoint=http://localhost:8080;TaskHub=default;Authentication=None"
  }
}

Minimal Example

const { app } = require("@azure/functions");
const df = require("durable-functions");

// Activity
df.app.activity("sayHello", {
  handler: (city) => `Hello ${city}!`,
});

// Orchestrator
df.app.orchestration("myOrchestration", function* (context) {
  const result1 = yield context.df.callActivity("sayHello", "Tokyo");
  const result2 = yield context.df.callActivity("sayHello", "Seattle");
  return `${result1}, ${result2}`;
});

// HTTP Starter
app.http("HttpStart", {
  route: "orchestrators/{orchestrationName}",
  methods: ["POST"],
  authLevel: "function",
  extraInputs: [df.input.durableClient()],
  handler: async (request, context) => {
    const client = df.getClient(context);
    const instanceId = await client.startNew(request.params.orchestrationName);
    return client.createCheckStatusResponse(request, instanceId);
  },
});

Workflow Patterns

Fan-Out/Fan-In

df.app.orchestration("fanOutFanIn", function* (context) {
  const cities = ["Tokyo", "Seattle", "London", "Paris", "Berlin"];

  // Fan-out: schedule all activities in parallel
  const tasks = cities.map((city) => context.df.callActivity("sayHello", city));

  // Fan-in: wait for all to complete
  const results = yield context.df.Task.all(tasks);
  return results;
});

Human Interaction

df.app.orchestration("approvalWorkflow", function* (context) {
  yield context.df.callActivity("sendApprovalRequest", context.df.getInput());

  // Wait for approval event with timeout
  const expiration = new Date(context.df.currentUtcDateTime);
  expiration.setDate(expiration.getDate() + 3);

  const approvalTask = context.df.waitForExternalEvent("ApprovalEvent");
  const timeoutTask = context.df.createTimer(expiration);

  const winner = yield context.df.Task.any([approvalTask, timeoutTask]);

  if (winner === approvalTask) {
    return approvalTask.result ? "Approved" : "Rejected";
  }
  return "Timed out";
});

Orchestration Determinism

❌ NEVER ✅ ALWAYS USE
new Date() context.df.currentUtcDateTime
Math.random() Pass random values from activities
setTimeout() context.df.createTimer()
Direct I/O, HTTP, database context.df.callActivity()

Replay-Safe Logging

df.app.orchestration("myOrchestration", function* (context) {
  if (!context.df.isReplaying) {
    console.log("Started");  // Only logs once, not on replay
  }
  const result = yield context.df.callActivity("myActivity", "input");
  return result;
});

Error Handling & Retry

df.app.orchestration("workflowWithRetry", function* (context) {
  const retryOptions = new df.RetryOptions(5000, 3); // firstRetryInterval, maxAttempts
  retryOptions.backoffCoefficient = 2.0;
  retryOptions.maxRetryIntervalInMilliseconds = 60000;

  try {
    const result = yield context.df.callActivityWithRetry(
      "unreliableService",
      retryOptions,
      context.df.getInput()
    );
    return result;
  } catch (ex) {
    context.df.setCustomStatus({ error: ex.message });
    yield context.df.callActivity("compensationActivity", context.df.getInput());
    return "Compensated";
  }
});

Durable Task SDK (Non-Functions)

For applications running outside Azure Functions (containers, VMs, Azure Container Apps, Azure Kubernetes Service):

const { createAzureManagedWorkerBuilder, createAzureManagedClient } = require("@microsoft/durabletask-js-azuremanaged");

const connectionString = "Endpoint=http://localhost:8080;Authentication=None;TaskHub=default";

// Activity
const sayHello = async (_ctx, name) => `Hello ${name}!`;

// Orchestrator
const myOrchestration = async function* (ctx, name) {
  const result = yield ctx.callActivity(sayHello, name);
  return result;
};

async function main() {
  // Worker
  const worker = createAzureManagedWorkerBuilder(connectionString)
    .addOrchestrator(myOrchestration)
    .addActivity(sayHello)
    .build();

  await worker.start();

  // Client
  const client = createAzureManagedClient(connectionString);
  const instanceId = await client.scheduleNewOrchestration("myOrchestration", "World");
  const state = await client.waitForOrchestrationCompletion(instanceId, true, 30);
  console.log("Output:", state.serializedOutput);

  await client.stop();
  await worker.stop();
}

main().catch(console.error);

Source: SKILL.md on GitHub

1 alert8d4 checks · Risk SAFE
  • Gen Agent Trust Hub8d

    The azure-prepare skill provides a comprehensive environment for preparing Azure applications for deployment. It focuses on generating infrastructure-as-code and deployment configuration while strictly enforcing security best practices like managed identity usage and secret management via Key Vault. No malicious patterns or security risks were identified.

  • Socket8d

    5 alerts: gptAnomaly, gptSecurity

  • Snyk8d

    Risk: LOW · No issues

  • Runlayer6mo

    68/196 files flagged

Signed by skilld at 7995851. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 3 weeks ago
metadata
{
  "author": "Microsoft",
  "version": "0.0.0-placeholder"
}
  • Infrastructure
  • azure
  • bicep
  • terraform
  • deployment
  • docker
  • functions
  • app-service

README badge

README badge for microsoft/github-copilot-for-azure/azure-prepare

Prepares Azure applications for deployment by generating infrastructure templates (Bicep or Terraform), azure.yaml configuration, and Dockerfiles. Covers new app creation, modernization, and hosting on App Service, Container Apps, or Functions—but excludes Python App Service deployments, copilot SDK apps, and cross-cloud migrations which have dedicated skills.

Generated from the current SKILL.md.

Does this skill handle Python App Service deployments?
No. Use the python-appservice-deploy skill instead for Python code-only App Service deploys.
Can I use this skill for cross-cloud migration?
No. This skill is for Azure-native preparation. Use azure-cloud-migrate for migrations from AWS, GCP, or other clouds.
Does this skill support Copilot SDK apps?
No. Use azure-hosted-copilot-sdk for apps with @github/copilot-sdk or CopilotClient.
What infrastructure templates does this skill support?
Azure Developer CLI (azd), Bicep, Terraform, and Azure CLI. The skill creates infrastructure code, Dockerfiles, and configuration files—actual deployment execution is handled by the azure-deploy skill.
Does this skill delete existing project files?
No. When adding features to existing projects, it modifies files rather than deletes them. It never removes the project or workspace directory itself.

Generated from the current SKILL.md. These answers refresh after source changes.