메뉴
"Crea cluster Kind con 1 control-plane y 5 workers (3 GPU, 2 CPU). Valida GPUs con pods nvidia-smi. USA cuando el usuario mencione 'crear cluster kind', 'kind cluster', 'kubernetes gpu', 'test gpu', 'nvidia-smi', o 'kind'."
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Workflow for maintaining and building the wpipe VS Code extension. Use when working in 'editors/vscode/' to compile, package, or update the extension.
Handles core library maintenance for wpipe, including example validation (pytest), concurrency fixes, version bumping, PyPI publication, and repository synchronization. Use when making changes to the 'wpipe/' directory or preparing a new release.
"Audita código Python completamente: Pylint ≥9.0, tests con coverage ≥85%, reportes HTML/PDF profesionales. USA esta skill cuando el usuario mencione 'auditar código', 'quality', 'coverage', 'tests', 'pylint', 'pytest coverage', o 'generar reporte de calidad'. Genera REPORT.md, REPORT.html y REPORT.pdf."
Revisa código Python automáticamente ANTES de presentarlo. Refactoriza, mejora calidad y genera Security Report estilo pentest. USA esta skill cuando: el usuario pida 'revisar código', 'analizar código', mencione 'seguridad', 'vulnerabilidades', o siempre ANTES de presentar código. MIENTRAS el agente escriba código, debe aplicar esta skill.
Convierte documentos (DOCX, PDF, Excel, ODT) a Markdown. USA cuando el usuario mencione 'convertir a markdown', 'DOCX a md', 'PDF a texto', 'Excel a tabla', 'migrar documentos', o 'extraer contenido de documento'.
Genera documentación completa: Sphinx (HTML) + LaTeX (PDF) + Technical Whitepaper. USA los PNGs de readme-makefile. Invoca sphinx-docs antes de PDF. USA cuando el usuario mencione 'generar PDF', 'documentación LaTeX', 'whitepaper', 'technical docs', o 'generar documentación completa'.
| name | kind-cluster |
| description | "Crea cluster Kind con 1 control-plane y 5 workers (3 GPU, 2 CPU). Valida GPUs con pods nvidia-smi. USA cuando el usuario mencione 'crear cluster kind', 'kind cluster', 'kubernetes gpu', 'test gpu', 'nvidia-smi', o 'kind'." |
Crear cluster Kind con:
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
name: gpu-cluster
nodes:
- role: control-plane
- role: worker
labels:
worker: worker1
resource: GPU
- role: worker
labels:
worker: worker2
resource: GPU
- role: worker
labels:
worker: worker3
resource: GPU
- role: worker
labels:
worker: worker4
resource: CPU
- role: worker
labels:
worker: worker5
resource: CPU
docker execapiVersion: v1
kind: Pod
metadata:
name: gpu-test-worker1
spec:
restartPolicy: Never
nodeSelector:
worker: worker1
containers:
- name: nvidia-smi
image: nvidia/cuda:12.2.0-base-ubuntu22.04
command: ["nvidia-smi"]
# Crear cluster
kind create cluster --config=kind-config.yaml
# Verificar nodos
kubectl get nodes --show-labels
# Validar GPU en cada worker
kubectl apply -f test-worker1.yaml
kubectl logs gpu-test-worker1
| worker_name | labels | available_gpu | result_by_pod | notes |
|---|---|---|---|---|
| worker1 | worker=worker1, resource=GPU | Yes/No | Success/Failed | - |
| worker2 | worker=worker2, resource=GPU | Yes/No | Success/Failed | - |
| worker3 | worker=worker3, resource=GPU | Yes/No | Success/Failed | - |
| worker4 | worker=worker4, resource=CPU | N/A | Success/Failed | CPU only |
| worker5 | worker=worker5, resource=CPU | N/A | Success/Failed | CPU only |