.claude/skills/azure-pipelines/SKILL.md
Skill para gestión de CI/CD con Azure Pipelines via MCP
npx skillsauth add gonzalezpazmonica/pm-workspace azure-pipelinesInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Skill para gestión de CI/CD con Azure Pipelines via MCP. Léela antes de cualquier operación con pipelines.
Azure Pipelines gestiona builds y releases de los proyectos.
PM-Workspace accede via el MCP azure-devops (ya declarado en mcp.json).
MCP tools de pipelines disponibles:
get_build_definitions — listar pipelines del proyectoget_builds — listar builds (filtrar por pipeline, branch, status)get_build_status — estado de una build específicaget_build_logs — logs de una build (timeline + contenido)run_pipeline — ejecutar una pipeline (requiere confirmación PM)preview_pipeline_run — preview de qué se ejecutará (sin ejecutar)create_pipeline — crear pipeline desde YAMLlist_artifacts — artefactos de una builddownload_artifacts — descargar artefactosUsa el mismo PAT de Azure DevOps configurado en pm-config.md.
Scopes requeridos: Build R/W (además de los existentes).
# El MCP lee el PAT del entorno automáticamente
# Verificar: el PAT debe tener scope "Build (Read & Execute)"
preview_pipeline_runoutput/artifacts/{proyecto}/{build-id}/MCP: get_build_definitions(project=PROYECTO)
→ Para cada definition: get_builds(project, definitionId, top=5)
→ Calcular: % éxito, duración media, trends
MCP: preview_pipeline_run(project, pipelineId, branch)
→ Mostrar resumen al PM → Confirmar
→ MCP: run_pipeline(project, pipelineId, branch, variables)
→ MCP: get_build_status(project, buildId) [polling si --watch]
1. Seleccionar template (references/yaml-templates.md)
2. Adaptar al proyecto (lenguaje, tests, deploy targets)
3. MCP: preview_pipeline_run → validar YAML
4. Confirmar con PM
5. MCP: create_pipeline(project, name, yamlPath, repository)
| Build Status | Significado | Acción PM |
|---|---|---|
| succeeded | Build OK | Ninguna |
| partiallySucceeded | Warnings o tests flaky | Revisar logs |
| failed | Error en build/test | Investigar + crear Bug si recurrente |
| canceled | Cancelado por usuario/timeout | Verificar causa |
| inProgress | Ejecutándose | Esperar o monitorizar |
| notStarted | En cola | Normal si hay carga |
Los pipelines multi-stage siguen el patrón:
Build → Test → Deploy DEV (auto) → Deploy PRE (approval) → Deploy PRO (approval)
Ver references/stage-patterns.md para patrones detallados.
Gates de aprobación:
references/yaml-templates.mdreferences/stage-patterns.md@docs/rules/domain/environment-config.md@.claude/mcp.jsontesting
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
tools
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
tools
Sistema proactivo de bienestar individual
development
Search the web to resolve context gaps — documentation, versions, CVEs, best practices. Auto-starts SearxNG Docker if available, falls back to WebSearch.