What problem does it solve? Running a controlled pressure test against a Huawei Cloud CCE workload requires coordinating cluster discovery, Kubernetes route verification, traffic generation, and evidence collection across multiple tools, and mistakes can affect production traffic or create billable resources. This Skill orchestrates that entire workflow with strict approval gates and produces an evidence-backed bottleneck report. ## Core Features & Use Cases - End-to-end pressure test orchestration: Discovers CCE clusters with hcloud, verifies routes and RBAC with the kubectl-cce plugin, and generates traffic with local k6 or an approved in-cluster k6 Job. - Staged traffic with safety gates: Enforces smoke-before-load phases, explicit approval for any mutation or traffic, stop conditions, and cleanup rules that never touch user resources. - Bottleneck analysis and reporting: Correlates k6 latency/error results with Pod metrics, HPA behavior, Events, and optional ELB/VPC evidence into a structured Markdown report with ranked findings. - Use Case: Ask the agent to run a ramp load test against an Ingress-fronted Deployment in cluster cn-north-4, and it will preflight the route, run a smoke phase, execute the approved k6 test, and report whether the bottleneck is the application, HPA lag, node capacity, or the ELB layer. ## Quick Start Ask the agent to run a smoke pressure test with k6 against your CCE workload by providing the region, cluster ID, namespace, workload name, and target URL.
