Cloud, CI/CD, monitoring, and deployment (6.8K skills)
Secure credentials used in CI/CD pipelines by selecting a secrets backend, limiting access, and integrating secret retrieval into builds or runtime workflows. Get guidance on rotation, audit logging, and avoiding hardcoded secrets in source control.
Plan and improve machine learning pipelines, experiment tracking, and model registries using tools such as MLflow and Kubeflow. Get actionable guidance for automating training, deployment, and monitoring across cloud platforms.
Create Kubernetes YAML for Deployments, Services, ConfigMaps, Secrets, and persistent storage. The skill guides resource limits, health checks, security contexts, naming, and configurations for multiple environments so workloads have clear, production-oriented manifests.
Use this skill to investigate production failures across application or infrastructure components. It guides the process from examining errors and isolating a root cause to implementing a focused fix and checking recovery with regression tests, with a playbook available for detailed patterns.
Compare AWS, Azure, and GCP services to shape a multi-cloud design for specific workloads. The skill helps evaluate provider choices, cloud-agnostic patterns, migration steps, disaster recovery, and cost considerations when combining or moving between cloud platforms.
Set service reliability targets by defining SLOs, the goals for service performance, and SLIs, the measures used to track them. Plan error budgets, dashboards, alerts, and reporting so teams can balance reliability with feature delivery using service data.
Coordinate a structured response to production incidents, from assessing severity and user impact through investigation, mitigation, recovery, communication, and follow-up. The workflow organizes findings and decisions so teams can restore service and identify improvements.
Plan Kubernetes security controls for network access, pod settings, and user permissions. The skill provides examples for NetworkPolicy, Pod Security Standards, and RBAC, helping define network boundaries and grant only the access workloads and users need.
Create incident runbooks that organize detection, triage, mitigation, recovery, verification, and communication into clear steps. The templates cover service outages and database incidents, with severity levels and escalation paths to guide responders during active issues.
Configure mutual TLS so both services authenticate each other before communicating. Find configuration examples for Istio, Linkerd, cert-manager, and SPIFFE/SPIRE, plus certificate rotation guidance and commands for investigating handshake and policy issues.
Set Istio policies for service routing, load balancing, retries, timeouts, and circuit breakers, which stop sending traffic to unhealthy services. Create canary rollouts, mirror traffic for testing, and configure gateways with practical resource examples and debugging commands.
Plan and troubleshoot cloud networks across AWS, Azure, and GCP. The skill helps with secure connectivity, service meshes, certificates, load balancing, CDN performance, monitoring, and network automation, with practical guidance for diagnosing connection and performance problems.
Plan Kubernetes platforms across cloud providers, including cluster layouts, GitOps delivery, progressive rollouts, security, and observability. GitOps keeps the desired system configuration in version control and reconciles changes automatically.
Use this skill to plan or improve monitoring for applications and infrastructure. It covers metrics, dashboards, distributed traces, log aggregation, and alerting, and outlines an architecture and implementation plan for gaining visibility into system health and performance.
Set up Linkerd service mesh patterns for Kubernetes, including automatic mutual TLS, access policies, canary traffic splits, retries, and timeouts. It provides configuration examples and commands for checking, monitoring, and debugging mesh deployments.
Use this skill to plan production monitoring, logs, and distributed tracing, which follows requests across services. It helps define reliability indicators and targets, build useful alerts, and investigate performance or reliability issues while reducing alert noise.
Prepare on-call handoffs that preserve incident context across shifts. Organize active incidents, investigations, recent changes, known issues, upcoming events, and escalation paths into practical summaries, meeting agendas, and checklists.
Design GitLab CI/CD pipelines that organize builds, tests, and deployments into stages. Use caching and GitLab Runners to manage automation, create environment-specific deployments, and add security scanning or Terraform workflows.
Set up Git based delivery for Kubernetes with ArgoCD or Flux. The skill helps organize configuration repositories, define automatic sync and environment promotion, handle secrets, and plan rollbacks or progressive releases such as canary deployments.
Create and organize Helm charts to package Kubernetes applications and manage reusable deployment templates. Use it to support multiple environments, prepare charts for distribution, and follow conventions when setting up chart repositories and deployments.
Create GitHub Actions workflows to automate testing, builds, and deployments. Use patterns for matrix builds, Docker publishing, Kubernetes deployment, reusable workflows, security scans, and production approval gates to structure development pipelines.
Analyze production incidents and recurring errors across services using error reports, logs, and traces. Narrow down causes with targeted investigation, then propose evidence-based fixes, checks, preventive measures, and improvements to observability and error handling.
Investigate errors by finding patterns, stack traces, and anomalies in logs and code. Correlate events across services and time periods to form evidence-based root-cause hypotheses, identify likely code locations, and suggest queries to detect recurrence.
Design Grafana dashboards for application, infrastructure, and business metrics. Get guidance and configuration examples for panels, variables, alerts, and provisioning, including service health and resource monitoring views that make operational data easier to inspect.