.claude/skills/predictive-analytics/SKILL.md
Fórmulas de predicción sprint, Monte Carlo simplificado y flow metrics
npx skillsauth add gonzalezpazmonica/pm-workspace predictive-analyticsInstall 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.
Entrada: velocity_history, items_remaining (pts), sprint_length, current_day
Algoritmo: for i=1 to 1000: sim_velocity = random(velocity_history); sprints_needed = ceil(remaining / velocity); completion_date = current_day + (sprints_needed * sprint_length); results.append(date)
Salida: percentiles P50, P70, P85, P95 (fechas posibles)
Interpretación:
Cálculo: P_X = percentile(results, X%) | Ejemplo: P70 = results[700] de 1000
Audiencia: Equipo interno→P50, Product Owner→P85, Ejecutivos→P95, Cliente→P70-P85
Advertencia: Nunca comunicar P50 como promesa sin contexto de variabilidad
Fórmula: (Active Time / Total Elapsed Time) × 100
Interpretación:
70%: excelente
Ejemplo: Item 13 días total | 7 días Active + 1 día Review = 8 activos | 8/13 = 61.5%
Agregado: Team FE = promedio(FE todos items completados en período)
Fórmula:
cycle_time_avg = promedio(últimos 20 items completados)
age = hoy - fecha_entrada_estado
ROJO: age > 2.0 × cycle_time_avg → investigar bloqueo
AMBER: age > 1.5 × cycle_time_avg → verificar progreso
VERDE: age ≤ 1.5 × cycle_time_avg → monitorear normal
Ejemplo: cycle_time=5d | ROJO>10d, AMBER>7.5d, VERDE≤7.5d
Entrada: items_completed_per_week (últimas 4-8 semanas)
Linear Regression: slope = cov(x,y) / var(x)
Requisito: mínimo 4 data points, ignorar outliers
WIQL - Velocity por Sprint:
SELECT [System.Id], [System.Title], [Microsoft.VSTS.Scheduling.StoryPoints]
WHERE [System.TeamProject] = @project AND [System.IterationPath] = @iteration AND [System.State] = 'Closed'
WIQL - State Transitions (Lead/Cycle Time):
SELECT [System.Id], [System.ChangedDate], [System.State]
WHERE [System.TeamProject] = @project AND [System.ChangedDate] >= @startDate
WIQL - WIP Aging:
SELECT [System.Id], [System.Title], [System.State], [System.StateChangeDate]
WHERE [System.State] IN ('Active', 'In Progress')
| Técnica | Limitación | |---------|-----------| | Monte Carlo | No considera cambios de equipo; regenerar si hay alta rotación | | Flow Metrics | Requiere logging consistente de transiciones de estado | | Velocity Trend | Mínimo 6 sprints para anomalía detection; vacaciones distorsionan | | Factores no-cuantitativos | Moral, requisitos calidad, acoplamiento técnico no se miden |
testing
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.