What problem does it solve?
Infrastructure-as-code and cloud-native configuration reviews are often ad-hoc and inconsistent, leading to missed critical security, reliability, and cost issues across Terraform, Kubernetes, Docker, Helm, gateways, and service meshes. This Skill standardizes the review process to ensure common misconfigurations are caught early, reducing production risk and technical debt.
Core Features & Use Cases
- Multi-surface coverage: Reviews Terraform/OpenTofu, Kubernetes manifests, Dockerfiles, Helm charts, north-south gateways (Envoy, Ingress, Gateway API), and east-west service meshes (Linkerd, Istio) in a single unified workflow.
- Integrated static analysis: Runs industry-standard linters and security scanners (tflint, tfsec, kubeconform, hadolint, etc.) when available, and interprets their output in context to avoid false positives.
- PR and follow-up support: Links findings directly to GitHub PR file lines for easy tracking, and supports re-evaluation of prior findings against updated code in follow-up reviews.
- Use Case: A team preparing a Kubernetes service deployment can run this review to catch missing resource limits, insecure security contexts, unauthenticated gateway routes, and misconfigured service mesh policies before they reach production.
Quick Start
Use the review-infrastructure skill to review all infrastructure-as-code files in the current repository for security, reliability, and best practice gaps.