Container and Orchestration Skills

docker-production-best-practices

What the skill prevents

kubernetes-manifests-and-helm

Key Kubernetes patterns

k8s-troubleshooting-runbooks

CI/CD Pipeline Skills

github-actions-workflows

gitlab-ci-pipelines

ci-security-scanning

Infrastructure as Code Skills

terraform-aws-modules

terraform-gcp-modules

pulumi-typescript-infra

Monitoring and Observability Skills

prometheus-grafana-observability

Example PromQL patterns the skill generates

datadog-apm-logging

Building Your DevOps Skill Stack

AWS with GitHub Actions

GCP with GitLab CI

Browse the Skills Directory

Built a Skill? List It on the Bazaar DevOps work is repetitive, configuration-heavy, and unforgiving of small mistakes. A missing indent in a YAML file, a wrong IAM permission, a misconfigured health check — these tiny errors cause outages and burn hours of debugging time. OpenClaw skills give your AI agent the deep infrastructure knowledge needed to get these details right the first time. This guide covers the most valuable OpenClaw skills for DevOps engineers and platform teams. Every skill is available in the OpenClaw Bazaar skills directory. Installs: 56,400 | Rating: 95% This is the most-installed DevOps skill on the Bazaar. It teaches your agent to generate Dockerfiles that follow production best practices: multi-stage builds, minimal base images, proper layer caching, non-root users, and health checks. It also covers Docker Compose for local development environments with service dependencies, volumes, and network configuration. Without this skill, agents tend to generate single-stage Dockerfiles with latest tags, running as root, with no .dockerignore file. The difference in image size and security posture is dramatic. Installs: 44,200 | Rating: 93% Kubernetes YAML is verbose and easy to misconfigure. This skill teaches your agent to generate correct Deployments, Services, Ingresses, ConfigMaps, Secrets, and Jobs. More importantly, it knows the production details: resource requests and limits, pod disruption budgets, rolling update strategies, and proper liveness and readiness probes. The skill also covers Helm chart creation and management — writing templates with proper value injection, creating reusable chart libraries, and structuring values.yaml files for multi-environment deployments. Installs: 18,700 | Rating: 91% When something goes wrong in a cluster, the last thing you want is an agent that gives generic advice. This skill contains runbook-style instructions for diagnosing common Kubernetes failures: CrashLoopBackOff, OOMKilled, ImagePullBackOff, pending pods, and networking issues. Your agent generates specific kubectl commands for diagnosis and suggests fixes based on the symptoms. Installs: 51,800 | Rating: 94% GitHub Actions is where most teams run their CI/CD, and the workflow syntax is deceptively complex. This skill teaches your agent to write workflows that use proper job dependencies, matrix builds, caching strategies, environment secrets, and reusable workflows. It covers both common patterns and advanced features like composite actions and workflow dispatch events. A typical improvement: without this skill, an agent might generate a workflow that installs dependencies on every job. With it, the agent uses actions/cache for dependency caching, sets up proper needs dependencies between jobs, and uses matrix strategies to test across multiple versions in parallel. Installs: 23,500 | Rating: 89% For teams on GitLab, this skill covers .gitlab-ci.yml configuration including stages, job dependencies, rules-based triggering, Docker-in-Docker builds, caching with cache:key, and multi-project pipeline orchestration. It also handles GitLab-specific features like environments, review apps, and DAST scanning integration. Installs: 19,300 | Rating: 90% Shifting security left means integrating scanning into your pipelines. This skill teaches your agent to configure SAST tools (Semgrep, CodeQL), dependency scanning (Dependabot, Snyk, Trivy), secret detection (Gitleaks, TruffleHog), and container image scanning. It generates pipeline steps that fail builds on critical vulnerabilities without creating so much noise that developers ignore the results. Installs: 47,300 | Rating: 94% Terraform is the dominant IaC tool, and this skill covers it thoroughly. Your agent learns to write modular Terraform configurations with proper state management, use data sources instead of hardcoded values, implement remote backends with state locking, and structure code with workspaces or directory-based environments. The skill is particularly strong on AWS — covering VPC design, IAM policies with least privilege, RDS configuration, ECS/Fargate task definitions, S3 bucket policies, CloudFront distributions, and ALB target group configuration. It prevents common mistakes like overly permissive security groups or S3 buckets without encryption. Installs: 21,600 | Rating: 91% The GCP equivalent of the AWS skill above. Covers Google Cloud-specific Terraform resources: GKE cluster configuration, Cloud Run service definitions, Cloud SQL instances, VPC networks, IAM bindings (the notoriously confusing google_project_iam_binding versus google_project_iam_member), and Cloud Build triggers. Installs: 14,800 | Rating: 88% For teams that prefer writing infrastructure in TypeScript instead of HCL, this skill covers Pulumi patterns: component resources, stack references, configuration management, and provider setup. It generates type-safe infrastructure code with proper output handling and resource dependencies. Installs: 28,900 | Rating: 92% Monitoring configuration is tedious but critical. This skill teaches your agent to write Prometheus scrape configurations, alerting rules with proper thresholds, and Grafana dashboard JSON. It covers PromQL queries for common metrics — request rate, error rate, latency percentiles, and resource utilization — and generates alerts that are actionable rather than noisy. Installs: 16,400 | Rating: 89% For Datadog users, this skill covers APM instrumentation, custom metric submission, log pipeline configuration, and monitor creation. It teaches your agent to set up distributed tracing, create meaningful dashboards, and write monitor queries that catch real problems without alert fatigue. The right combination depends on your infrastructure. Here are two common setups. DevOps skills are some of the highest-impact skills on the Bazaar because infrastructure code is both critical and easy to get subtly wrong. For tips on evaluating skills before installing, read our guide to finding the right skill. Find the right skill for your workflow. The OpenClaw Bazaar skills directory has over 2,300 community-rated skills — searchable, sortable, and free to install. If you have built a skill that others would find useful, publish it on the Bazaar. Reach thousands of developers and get feedback from the community. Learn how to publish → Templates let you quickly answer FAQs or store snippets for re-use. Hide child comments as well For further actions, you may consider blocking this person and/or reporting abuse