Typical Workflow Example

This document provides a concrete example of the end-to-end process for generating a complex Azure Pipeline.

Example Request

User request: “Create a CI/CD pipeline for a Node.js app with Docker deployment to AKS”

Step-by-Step Process

1. UNDERSTAND  — Node.js build/test → Docker image → push to ACR → deploy to AKS (staging + prod)
2. READ        — references/yaml-schema.md, tasks-reference.md, best-practices.md
                 assets/examples/multi-stage-cicd.yml, kubernetes-deploy.yml
3. LOOKUP      — Context7 or WebSearch for any tasks not covered locally
4. GENERATE    — Stage 1: NodeTool@0 + npm ci (cached) + build + test + PublishTestResults
                 Stage 2: Docker@2 buildAndPush (tag with BuildId + latest)
                 Stage 3: Deployment job → KubernetesManifest@0 → health check
5. VALIDATE    — Invoke devops-skills:azure-pipelines-validator; fix errors; re-validate
6. PRESENT     — Validated pipeline + stage explanation + setup notes (service connections, environments)

Workflow Components Explained

1. UNDERSTAND

Break down the user’s request into technical requirements:

Build: Node.js application needs runtime setup, dependency installation, compilation
Test: Run automated tests and publish results
Package: Create Docker container image
Publish: Push image to Azure Container Registry (ACR)
Deploy: Deploy to Azure Kubernetes Service across multiple environments

2. READ

Load relevant documentation and examples:

Core schema and task references for syntax accuracy
Best practices for security, performance, naming conventions
Example files matching the pipeline type (multi-stage, Kubernetes)

3. LOOKUP

Use external documentation sources when needed:

Tasks not covered in local references
Version-specific behavior questions
Troubleshooting specific errors

4. GENERATE

Create the pipeline following:

Multi-stage structure with proper dependencies
Security best practices (service connections, no hardcoded secrets)
Performance optimizations (caching, parallelism)
Proper error handling and test result publishing

5. VALIDATE

Critical step: Run validation before delivery:

# Invoke validator skill
devops-skills:azure-pipelines-validator

Fix any reported issues and re-validate until clean.

6. PRESENT

Deliver complete solution with:

Validated pipeline YAML
Explanation of each stage’s purpose
Setup instructions (service connections, environment configuration)
Any additional configuration needed in Azure DevOps