1. azure-deploy
Deploys to Azure using azd, Terraform, and native Azure deployment commands with structured validation and recovery steps built in. Covers resource groups, app services, and container instances. Every deployment runs pre-flight checks and post-deployment verification so you catch failures before they become incidents.
2. azure-validate
Validates Azure configurations before they hit production. The skill audits ARM templates, Bicep files, and azd configs for common misconfigurations — open NSG rules, missing diagnostics settings, weak RBAC assignments — and flags them with remediation steps. Run it as a pre-deploy gate in CI or on demand before any significant infra change.
3. superpowers:security-review
An OWASP-aligned review of infrastructure configs and code changes. The skill checks for secrets in code, insecure defaults, privilege escalation paths, and missing input validation — running through a structured checklist every time. Ideal before merging to main or promoting from staging to production. It produces a prioritized findings report, not just a raw scan dump.
4. kubernetes-deploy
Generates production-ready Kubernetes manifests including Deployments, Services, Ingress rules, HPAs, and ConfigMaps. It follows the CIS Kubernetes Benchmark out of the box — adding readiness and liveness probes, resource limits, and RBAC rules without you having to remember every YAML key. Works with AKS, EKS, GKE, and any vanilla cluster, and verifies rollout health after each apply.
5. railway-metrics
Pulls Railway.app monitoring data and surfaces anomalies fast. The skill analyzes memory usage, CPU spikes, request latency, and deploy history to give you a health snapshot without clicking through dashboards. Useful for rapid triage during an incident and for building a baseline understanding of normal behavior before you need it under pressure.
6. azure-cost-optimization
Identifies Azure spend waste and rightsizes resources. The skill analyzes your Azure Cost Management data, flags idle or overprovisioned resources, and produces a ranked list of savings opportunities with the estimated monthly impact. Particularly useful at month-end reviews or before committing to reserved instances — it gives you the data to make a defensible recommendation rather than a gut-feel estimate.
7. superpowers:systematic-debugging
Structured root cause analysis for production incidents. Instead of random reboots and crossed fingers, this skill walks through a repeatable diagnostic framework — isolating the failure domain, gathering evidence, forming hypotheses, and verifying fixes before marking an incident resolved. The output is a structured incident write-up that doubles as your postmortem draft, saving time under pressure and improving your retrospectives.
8. github-pages
Automates GitHub Pages deployment and configuration. The skill handles branch setup, custom domain configuration, CNAME management, and workflow file generation so you stop manually tweaking repository settings. Useful for internal docs, release notes sites, and any static asset that lives on Pages — especially in organizations where many repos need consistent Pages configs applied at once.
How to set up a DevOps skill pack
The most effective approach is to install all 8 skills in a single session and commit the skill folder to your repository so the whole team gets them automatically.
Install all 8 in one pass
Run these commands from your terminal. The -y flag skips confirmation prompts so you can chain them without interruption:
# Azure deploy + validate
npx skills add microsoft/azure-skills@azure-deploy -g -y
npx skills add microsoft/azure-skills@azure-validate -g -y
npx skills add microsoft/azure-skills@azure-cost-optimization -g -y
# Security and debugging
npx skills add obra/superpowers@security-review -g -y
npx skills add obra/superpowers@systematic-debugging -g -y
# Kubernetes and container deploys
npx skills add microsoft/azure-skills@kubernetes-deploy -g -y
# Monitoring and Pages
npx skills add davila7/claude-code-templates@railway-metrics -g -y
npx skills add alfredang/skills@github-pages -g -y
Commit skills to your repo
For team setups, install without the -g flag (project-level install) and commit the .claude/skills/ folder to your repository. This means everyone who clones the repo has the same skill set without running any install commands.
# Install at project level (no -g flag)
npx skills add microsoft/azure-skills@azure-deploy -y
npx skills add obra/superpowers@security-review -y
# ... repeat for each skill
# Commit the skill folder to source control
git add .claude/skills/
git commit -m "feat: add DevOps skill pack to repo"