evidra

Use this skill when the user asks you to perform DevOps operations — diagnosing, fixing, or managing infrastructure using kubectl, helm, terraform, or aws commands. This includes: running kubectl get/describe/logs/apply/delete/patch, helm install/upgrade/status/list, terraform plan/apply/destroy/import, aws CLI commands, or any infrastructure investigation and remediation workflow. Also trigger when the user mentions Evidra, evidence chains, prescribe/report protocol, reliability scorecards, or run_command. DO NOT trigger for: writing Dockerfiles, writing Ansible/Terraform code without executing it, CI/CD pipeline setup, explaining infrastructure concepts, or writing tests for infrastructure tools. The key distinction is OPERATING infrastructure (both reading and mutating) vs WRITING code.

Use this skill when the user asks you to perform DevOps operations — diagnosing, fixing, or managing infrastructure using kubectl, helm, terraform, or aws commands. This includes: running kubectl get/describe/logs/apply/delete/patch, helm install/upgrade/status/list, terraform plan/apply/destroy/import, aws CLI commands, or any infrastructure investigation and remediation workflow. Also trigger when the user mentions Evidra, evidence chains, prescribe/report protocol, reliability scorecards, or run_command. DO NOT trigger for: writing Dockerfiles, writing Ansible/Terraform code without executing it, CI/CD pipeline setup, explaining infrastructure concepts, or writing tests for infrastructure tools. The key distinction is OPERATING infrastructure (both reading and mutating) vs WRITING code.

Use this skill when the user asks you to EXECUTE infrastructure mutations — commands that create, modify, or delete real resources. This includes: kubectl apply/delete/patch/create/replace/scale/rollout restart, helm install/upgrade/uninstall/rollback, terraform apply/destroy/import, docker run/rm/stop/kill, docker-compose up/down, podman/nerdctl equivalents, or any IaC tool that changes live infrastructure state. Also trigger when the user mentions Evidra, evidence chains, prescribe/report protocol, or reliability scorecards. DO NOT trigger for: writing Dockerfiles, writing Ansible/Terraform code without executing it, CI/CD pipeline setup, kubectl get/describe/logs (read-only), helm list/status/template, terraform plan/show, explaining infrastructure concepts, or writing tests for infrastructure tools. The key distinction is EXECUTING mutations vs WRITING code or READING state.

Use this skill when the user asks you to EXECUTE infrastructure mutations — commands that create, modify, or delete real resources. This includes: kubectl apply/delete/patch/create/replace/scale/rollout restart, helm install/upgrade/uninstall/rollback, terraform apply/destroy/import, docker run/rm/stop/kill, docker-compose up/down, podman/nerdctl equivalents, or any IaC tool that changes live infrastructure state. Also trigger when the user mentions Evidra, evidence chains, prescribe/report protocol, or reliability scorecards. DO NOT trigger for: writing Dockerfiles, writing Ansible/Terraform code without executing it, CI/CD pipeline setup, kubectl get/describe/logs (read-only), helm list/status/template, terraform plan/show, explaining infrastructure concepts, or writing tests for infrastructure tools. The key distinction is EXECUTING mutations vs WRITING code or READING state.