Skip to main content

DevOps Strategy

Generate comprehensive DevOps strategy with CI/CD, Infrastructure as Code, and platform engineering.

Command

arckit devops "<project ID or platform>"

Description

Defines how software will be built, tested, deployed, and managed throughout its lifecycle. Establishes engineering practices, tooling, and automation for rapid, reliable delivery.

Arguments

  • Project ID or platform: Project identifier or platform description (e.g., ‘001’, ‘GitHub Actions on AWS’)

When to Use

  • After requirements and architecture diagrams
  • Before implementation begins
  • For platform engineering initiatives
  • CI/CD pipeline design
  • Infrastructure automation planning

Required Context

  • Requirements (ARC--REQ-.md) - MANDATORY
  • Architecture Principles (ARC-000-PRIN-*.md) - MANDATORY
  • Architecture Diagrams (DIAG) - Recommended
  • Technology Research (RSCH/AWSR/AZUR) - Recommended

DevOps Maturity Levels

LevelCharacteristicsDeployment Frequency
Level 1Manual builds, scripted deploysMonthly
Level 2CI automation, manual deploysWeekly
Level 3CI/CD automation, staging gatesDaily
Level 4Continuous deployment, feature flagsMultiple/day
Level 5GitOps, self-healing, platformOn-demand

Document Sections

Section 1: DevOps Overview

  • Strategic objectives
  • Maturity level (current and target)
  • Team structure (platform team, dev teams)
  • Key stakeholders

Section 2: Source Control Strategy

Repository Structure:
  • Monorepo vs multi-repo decision
  • Repository naming conventions
  • Access control and permissions
Branching Strategy:
  • GitFlow / Trunk-based / GitHub Flow
  • Flowchart diagram (NOT gitGraph - use flowchart)
  • Protected branches and merge rules
Code Review:
  • Required approvers
  • Review checklist
  • Automated checks (linting, tests)
Commit Conventions:
  • Format: type(scope): message
  • Types: feat, fix, docs, refactor, test, chore

Section 3: CI Pipeline Design

Pipeline Architecture:
  • Stages: Build → Test → Analyze → Package
  • Parallel job execution
  • Artifact versioning
Build Automation:
  • Language-specific build tools
  • Dependency management
  • Build caching
Testing Strategy:
  • Unit tests (80% coverage target)
  • Integration tests
  • E2E tests
  • Contract tests
Quality Gates:
  • Linting (ESLint, Pylint, etc.)
  • Formatting (Prettier, Black)
  • Code coverage threshold
  • Security scanning
Security Scanning:
  • SAST (Static Application Security Testing)
  • Dependency scanning (Snyk, Dependabot)
  • License compliance

Section 4: CD Pipeline Design

Deployment Pipeline:
  • Stages: Dev → Staging → Production
  • Environment promotion gates
  • Approval requirements
Deployment Strategies:
  • Blue-Green deployment
  • Canary deployment
  • Rolling deployment
  • Feature flags integration
Approval Gates:
  • Manual approval for production
  • Automated smoke tests
  • Compliance checks
Rollback Procedures:
  • Automated rollback triggers
  • Rollback execution steps
  • Rollback verification

Section 5: Infrastructure as Code

IaC Tool Selection:
  • Terraform / Pulumi / CloudFormation
  • Module/component structure
  • Reusable modules
State Management:
  • Remote state backend (S3, Terraform Cloud)
  • State locking
  • State file encryption
Secret Management:
  • Vault / AWS Secrets Manager / Azure Key Vault
  • Secret rotation policies
  • Application secret injection
Drift Detection:
  • Scheduled drift detection
  • Remediation workflow

Section 6: Container Strategy

Container Runtime:
  • Docker / containerd
  • Base image strategy
  • Multi-stage builds
Image Registry:
  • ECR / ACR / GCR / Docker Hub
  • Image tagging strategy
  • Image retention policies
Image Security:
  • Vulnerability scanning (Trivy, Clair)
  • Image signing (Cosign)
  • Minimal base images (Alpine, Distroless)

Section 7: Orchestration (if applicable)

Kubernetes/ECS/AKS/GKE:
  • Cluster architecture
  • Namespace strategy
  • Resource limits and quotas
  • Service mesh (Istio, Linkerd)
  • Ingress/networking
GitOps Tooling:
  • ArgoCD / Flux
  • Git as source of truth
  • Automated sync

Section 8: Environment Management

Environment Types:
  • Dev (feature branches)
  • Staging (pre-production)
  • Production (live)
Environment Parity:
  • Configuration as code
  • Infrastructure templates
  • Data management across environments
Ephemeral Environments:
  • PR preview environments
  • Auto-cleanup policies

Section 9: Developer Experience

Local Development:
  • Dev containers / devcontainers
  • Docker Compose for local stack
  • Local environment setup automation
Inner Loop Optimization:
  • Fast feedback loops
  • Hot reload / live reload
  • Local testing capabilities
Self-Service:
  • Platform API or portal
  • Service templates
  • Automated provisioning

Section 10: Observability Integration

Logging:
  • Centralized logging (ELK, Splunk, CloudWatch)
  • Structured logging format
  • Log retention policies
Metrics:
  • Prometheus / CloudWatch / Datadog
  • Custom metrics instrumentation
  • Dashboard provisioning as code
Tracing:
  • Jaeger / X-Ray / Zipkin
  • Distributed tracing
  • Span correlation
Alerting:
  • Alert rules as code
  • Escalation policies
  • On-call rotation

Section 11: DevSecOps

Shift-Left Security:
  • Security scanning in CI
  • Pre-commit hooks
  • Security training
Security Scanning Tools:
  • SAST: SonarQube, Semgrep
  • DAST: OWASP ZAP
  • SCA: Snyk, Dependabot
  • Container: Trivy, Clair
  • Infrastructure: Checkov, tfsec
Compliance as Code:
  • Policy as code (OPA, Sentinel)
  • Compliance scanning
  • Audit trails

Section 12: Release Management

Versioning:
  • Semantic Versioning (SemVer)
  • Automated version bumping
  • Git tags
Changelog:
  • Automated changelog generation
  • Conventional Commits
Release Process:
  • Release branches
  • Release notes
  • Hotfix process

Section 13: Platform Engineering (if applicable)

Internal Developer Platform:
  • Self-service portal
  • Golden paths / templates
  • Platform APIs
  • Service catalog

Section 14: UK Government Compliance (if applicable)

  • Cloud First (TCoP Point 5) implementation
  • Open standards (TCoP Point 4)
  • Secure by Design integration
  • Digital Marketplace compatibility

Section 15: Metrics & Improvement

DORA Metrics:
  • Deployment frequency
  • Lead time for changes
  • Mean time to recovery (MTTR)
  • Change failure rate
Engineering Metrics:
  • Build success rate
  • Test coverage
  • Code review time
Continuous Improvement:
  • Retrospectives
  • Metrics review cadence
  • Automation opportunities

Section 16: Traceability

  • Requirements to DevOps element mapping
  • NFR-P → Build speed targets
  • NFR-S → Auto-scaling
  • NFR-SEC → Security scanning

Output

Creates: projects/{project}/ARC-{PROJECT_ID}-DEVOPS-v1.0.md

Example

arckit devops "Payment gateway on AWS with GitHub Actions"
Output:
  • DevOps Maturity: Level 3 (target: Level 4)
  • Cloud Provider: AWS
  • Deployment: EKS (Kubernetes)
  • CI: GitHub Actions
  • IaC: Terraform
  • Container Registry: ECR
  • Orchestration: EKS
  • SAST: SonarQube
  • Target: Daily deployments
  • arckit finops - Cloud cost management strategy
  • arckit servicenow - Operational ITSM integration
  • arckit diagram - Deployment topology

Next Steps

After DevOps strategy:
  1. Set up source control repository
  2. Implement CI pipeline
  3. Provision infrastructure with IaC
  4. Configure CD pipeline
  5. Set up secret management
  6. Establish DORA metrics baseline
  7. Train team on platform

Key Principles

  1. Automation First - Automate everything possible
  2. Security Shift-Left - Security in CI, not just production
  3. Infrastructure as Code - No manual changes
  4. Developer Experience - Fast feedback, self-service
  5. Observability by Default - Logging, metrics, tracing from day one

Build docs developers (and LLMs) love