Cost driver identification — effort inductors, scope drivers, magnitude estimation, team composition modeling, risk-adjusted timeline ranges, service engagement sizing, consulting effort, automation ROI, and staffing model. Use when the user asks to "estimate effort", "identify cost drivers", "size the project", "plan team composition", "identify effort inductors", or mentions WBS, sizing, contingency, burn rate, PERT, Monte Carlo, or "Phase 4" cost work. NEVER produces final prices — produces drivers, ranges, and magnitude indicators with costing disclaimers.
From pmnpx claudepluginhub javimontano/mao-pm-apexThis skill is limited to using the following tools:
examples/README.mdexamples/sample-output.htmlexamples/sample-output.mdexamples/sample-output.pptx-spec.mdexamples/sample-output.xlsx-spec.mdprompts/metaprompts.mdprompts/use-case-prompts.mdreferences/body-of-knowledge.mdreferences/knowledge-graph.mmdreferences/state-of-the-art.mdSearches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Translates technical scope into effort drivers, magnitude indicators, team composition models, and risk-adjusted timeline ranges. Produces structured analysis of WHAT drives cost — not WHAT things cost. Every output carries explicit disclaimers separating cost identification from pricing decisions.
El costeo sin estructura es adivinación con formato de hoja de cálculo. Este skill impone disciplina analítica sobre la estimación: cada driver se identifica, cada magnitud se triangula, cada supuesto se documenta. No producimos precios — producimos la base de conocimiento que hace posible tomar decisiones financieras informadas.
Drivers, no precios. El valor de este skill no está en producir un número final — está en identificar TODO lo que compone ese número. Los drivers son la verdad; el precio es una decisión posterior.
Triangulación obligatoria. Un solo método de estimación es una opinión. Dos métodos son una hipótesis. Tres métodos convergentes son confianza. Siempre triangular.
Incertidumbre explícita. Rangos, no puntos. Escenarios, no certezas. El Cone of Uncertainty no es debilidad — es honestidad profesional que genera confianza.
NUNCA producir valores finales de costo, precio o tarifa. Este skill identifica CONDUCTORES de costo, INDUCTORES de esfuerzo, y NOCIONES DE MAGNITUD.
Costear ≠ Cobrar ≠ Ingresos. El costeo existe para entender qué cuestan las cosas — completamente desconectado de lo que se cobra. El revenue es una decisión comercial posterior e independiente. Este skill vive exclusivamente en el dominio del costeo.
Costear para la Excelencia. El propósito del costeo NO es solo presupuestar — es asegurar calidad, excelencia, y un "wow factor" de hospitalidad irracional. Cuando sabes lo que las cosas verdaderamente cuestan, puedes invertir apropiadamente en calidad. Costear bien = habilitar excelencia.
Margen de innovación = inversión en futuro. El 5% adicional no es contingencia — es la declaración de que la excelencia no es accidental. Es presupuesto deliberado para sorprender al cliente positivamente.
La diferencia:
| Este skill PRODUCE | Este skill NO produce |
|---|---|
| "Requiere 3 seniors + 2 mids × 18 meses" | "$1,200,000 USD" |
| "Licenciamiento enterprise tier de {vendor}" | "$45,000/año" |
| "Infra: 3 ambientes × cluster K8s + DB managed" | "$8,500/mes cloud" |
| "Magnitud: proyecto mediano-alto (100-200 FTE-meses)" | "Costo total: $2.3M" |
| "Driver principal: integración con 7 sistemas legacy" | "Integración costará $350K" |
Todo output DEBE incluir al pie:
DISCLAIMER DE COSTEO
═══════════════════
Este análisis identifica conductores de costo e inductores de esfuerzo.
NO constituye una cotización, presupuesto ni compromiso financiero.
Los valores finales requieren: (1) validación de tarifas vigentes,
(2) negociación comercial, (3) aprobación de alcance definitivo.
Costear ≠ Cobrar. Este documento informa lo primero.
Estimates narrow as projects progress. At concept phase: 0.25x-4x. After requirements: 0.67x-1.5x. After detailed design: 0.8x-1.25x. Communicate ranges, not points. Re-estimate at each phase gate.
Parse $1 as project/initiative name. Detect project context from repo.
Parameters:
{MODO}: piloto-auto (default) | desatendido | supervisado | paso-a-paso
{FORMATO}: markdown (default) | html | dual{VARIANTE}: ejecutiva (~40% — S1 scope + S4 drivers + S6 magnitude) | técnica (full 7 sections, default){TIPO_SERVICIO}: SDA (default) | QA | Management | RPA | Data-AI | Cloud | SAS | UX-Design
When {TIPO_SERVICIO} ≠ SDA, use service-appropriate decomposition:
| Service Type | Decomposition Units | Complexity Drivers |
|---|---|---|
| QA | Test suites, automation scripts, environments, test data | Test case count, tool complexity, integration points |
| Management | Workshops, sprints, ceremonies, coaching sessions, deliverables | Team size, methodology complexity, stakeholder count |
| RPA | Processes to automate, bots, integrations, exception handlers | Process steps, decision points, system integrations |
| Data-AI | Pipelines, models, dashboards, data products, governance policies | Data volume, model complexity, source count |
| Cloud | Workloads to migrate, environments, automation scripts, runbooks | Workload complexity (7R), dependency count, compliance needs |
| SAS | Positions to fill, ramp-up plans, knowledge transfer sessions | Role specialization, market scarcity, domain complexity |
| UX-Design | Research studies, wireframes, prototypes, design system components | User complexity, platform count, accessibility requirements |
COCOMO II applies to SDA only. For other service types, use:
Universal methods that apply to ALL service types:
| Service Type | Typical Team Composition |
|---|---|
| QA | QA Lead, Test Analysts, Automation Engineers, Performance Testers, Test Manager |
| Management | PM/Scrum Master, Delivery Manager, Agile Coach, Product Owner, UX Specialist |
| RPA | RPA Architect, RPA Developers, Process Analyst, BPMN Analyst, RPA Tester |
| Data-AI | Data Architect, Data Engineers, Data Scientists, ML Engineers, Analytics Engineers, Data Analyst |
| Cloud | Cloud Architect, DevOps Engineers, SREs, Cloud Engineers, DevSecOps Engineer |
| SAS | Talent Acquisition Lead, Technical Interviewer, Onboarding Specialist, Account Manager |
| UX-Design | UX Lead, UX Researcher, UI Designer, Interaction Designer, Accessibility Specialist |
NUEVA SECCIÓN — el corazón del skill evolucionado.
Identifica y clasifica TODOS los drivers de costo:
| Categoría | Drivers | Cómo Identificar |
|---|---|---|
| Personal | # FTEs, seniority mix, duración, ramp-up | Del WBS y modelo de equipo |
| Licenciamiento | Vendors, tiers (community/enterprise), periodicidad | Del stack tecnológico AS-IS y TO-BE |
| Infraestructura | Ambientes (dev/staging/prod), compute, storage, networking | De arquitectura y deployment |
| Herramientas | CI/CD, monitoring, testing, project management | De pipeline DevOps |
| Training | Capacitación en stack nuevo, certificaciones | De gap de skills del equipo |
| Migración | Volumen de datos, ventanas de migración, rollback | De modelo de datos y SLAs |
| Compliance | Auditorías, penetration testing, certificaciones | De regulación de industria |
| Contingencia | Known risks (10-15%), unknown-unknowns (15-25%) | Del risk register |
| Oportunidad | Costo de NO hacer: deuda acumulada, riesgo operacional | Del AS-IS |
| Service Type | Additional Drivers |
|---|---|
| QA | Test tool licenses (Tricentis, Tosca), test environment provisioning, test data management, ISTQB certification costs |
| Management | Certification costs (PMP, CSM, SAFe), workshop facilitation tools, travel/onsite presence, methodology licensing |
| RPA | Bot platform licenses (UiPath, AA, Power Automate), process mining tools, production bot orchestration infrastructure |
| Data-AI | Data platform licenses (Databricks, Snowflake), GPU compute for training, data labeling, model monitoring tools |
| Cloud | Cloud consumption (pay-as-you-go), migration tooling licenses, multi-cloud management, security tooling |
| SAS | Recruitment platform costs, background check costs, onboarding infrastructure, bench time (between assignments) |
| UX-Design | Design tool licenses (Figma, Sketch), usability testing platforms, research participant incentives, accessibility audit tools |
Por cada driver:
Diagramas requeridos:
No produce presupuestos. Produce marcos de magnitud.
| Decision | Enables | Constrains | When to Use |
|---|---|---|---|
| Bottom-up drivers | Granular, traceable | Time-consuming | Post-discovery |
| Top-down analogous | Fast magnitude | Less precise | Pre-discovery |
| Monte Carlo ranges | Explicit uncertainty | Needs 3-point estimates | Stakeholder comms |
| Phased funding | Risk mitigation | Slower start | High uncertainty |
| Scenario | Response |
|---|---|
| Client asks for final price | Redirect: "Este análisis identifica drivers. El pricing es decisión comercial separada." |
| Greenfield / no history | Reference-class forecasting. Wider ranges. Flag as high uncertainty. |
| Legacy modernization | +30-50% buffer. Parallel running as driver. |
| Multi-vendor | +15-25% communication overhead driver. |
| Regulatory-heavy | Compliance driver: +20-40% testing effort. |
| Format | Default | Description |
|---|---|---|
markdown | ✅ | Rich Markdown + Mermaid diagrams. Token-efficient. |
html | On demand | Branded HTML (Design System). Visual impact. |
dual | On demand | Both formats. |
Default output is Markdown with embedded Mermaid diagrams. HTML generation requires explicit {FORMATO}=html parameter.
Primary: 06_Cost_Drivers_{TIPO_SERVICIO}_{project}.md (o .html si {FORMATO}=html|dual) — Effort drivers, magnitude indicators, team model, timeline ranges, costing governance. Con disclaimer obligatorio.
Diagramas incluidos:
| Caso | Estrategia de Manejo |
|---|---|
| Cliente pide precio final | Redirigir: "Este analisis identifica drivers. El pricing es decision comercial separada." Producir drivers y magnitudes, nunca precios. |
| Greenfield sin historia | Reference-class forecasting. Rangos mas amplios. Flag como alta incertidumbre (Cone of Uncertainty en fase concepto: 0.25x-4x). |
| Legacy modernization | +30-50% buffer por complejidad no documentada. Parallel running como driver adicional. |
| Multi-vendor engagement | +15-25% communication overhead driver. Coordination costs explicitos en taxonomia. |
| Regulatory-heavy industry | Compliance driver: +20-40% testing effort. Auditoria y certificaciones como drivers separados. |
| Scope muy ambiguo (pre-discovery) | Solo top-down analogous estimation. Rangos amplios con Cone of Uncertainty. Re-estimar post-discovery. |
| Decision | Alternativa Descartada | Justificacion |
|---|---|---|
| Drivers y magnitudes, NUNCA precios | Producir presupuesto final | Costear y cobrar son dominios separados. El skill produce la base de conocimiento (drivers, inductores, magnitudes). El pricing es decision comercial posterior con variables externas. |
| Triangulacion obligatoria (3+ metodos) | Un solo metodo de estimacion | Un metodo es opinion. Dos son hipotesis. Tres convergentes son confianza. Divergencia >30% entre metodos es red flag que requiere investigacion. |
| Margen de innovacion 5% separado de contingencia | Solo contingencia, sin margen de innovacion | El 5% de innovacion es inversion deliberada en excelencia y hospitalidad irracional. No es contingencia (que se calcula aparte por riesgo). |
| Service-type specific sizing methods | COCOMO II para todo | COCOMO II aplica solo a SDA. QA, RPA, Cloud, Data-AI tienen unidades de sizing fundamentalmente diferentes (test cases, bots, workloads, pipelines). |
graph TD
subgraph Core["Conceptos Core"]
WBS["Scope Decomposition"]
SIZING["Sizing Methods"]
TEAM["Team Composition"]
DRIVERS["Cost Driver Taxonomy"]
TIMELINE["Risk-Adjusted Timeline"]
MAGNITUDE["Magnitude Framing"]
GOVERNANCE["Costing Governance"]
end
subgraph Inputs["Entradas"]
ASIS["AS-IS Analysis"]
TOBE["TO-BE Architecture"]
SCENARIO["Approved Scenario"]
TIPO["Service Type"]
end
subgraph Outputs["Salidas"]
REPORT["Cost Drivers Report"]
GANTT["Ramp-up Gantt"]
MINDMAP["Driver Taxonomy Mindmap"]
FLOWCHART["Magnitude Decision Tree"]
end
subgraph Related["Skills Relacionados"]
ASISSK["asis-analysis"]
TOBEARCHSK["architecture-tobe"]
SCENARIOSK["scenario-evaluation"]
FEASIBILITY["feasibility-assessment"]
end
ASIS --> WBS
TOBE --> DRIVERS
SCENARIO --> MAGNITUDE
TIPO --> SIZING
WBS --> SIZING
SIZING --> TEAM
TEAM --> DRIVERS
DRIVERS --> TIMELINE
TIMELINE --> MAGNITUDE
MAGNITUDE --> GOVERNANCE
GOVERNANCE --> REPORT
REPORT --> GANTT
REPORT --> MINDMAP
REPORT --> FLOWCHART
ASISSK -.-> WBS
TOBEARCHSK -.-> DRIVERS
SCENARIOSK -.-> MAGNITUDE
FEASIBILITY -.-> TIMELINE
Formato Markdown (default):
# Cost Drivers: {project} ({TIPO_SERVICIO})
## S1: Scope Decomposition & Effort Drivers
### WBS
### Feature Inventory
| Feature | Complexity | Effort Drivers | Dependencies |
...
## S2: Sizing Methods
### Triangulacion de Magnitud
| Metodo | Resultado (FTE-meses) | Confianza |
...
## S3: Team Composition Model
### Modelo por Fase (roles y cantidades, NO tarifas)
### Gantt de Ramp-up (Mermaid)
## S4: Cost Driver Taxonomy
### Mindmap (Mermaid)
| Categoria | Driver | Magnitud | Fase(s) | Owner |
...
## S5: Risk-Adjusted Timeline
### PERT: P50 / P80 / P95
## S6: Magnitude Framing
### Clasificacion: {micro|pequeno|mediano|grande|enterprise}
### Margen de Innovacion: 5%
## S7: Costing Governance
> DISCLAIMER DE COSTEO
Formato HTML (bajo demanda):
06_Cost_Drivers_{TIPO_SERVICIO}_{project}_{WIP}.html
HTML self-contained branded (Design System MetodologIA v5). Light-First Technical. Incluye cost driver taxonomy mindmap interactivo, magnitude framing visual, y Gantt de ramp-up del equipo. WCAG AA, responsive, print-ready.
Formato XLSX (bajo demanda):
Sheet 1: WBS — epic, feature, task, complexity, effort drivers, dependencies
Sheet 2: Sizing Triangulation — method, result (FTE-months), confidence, divergence
Sheet 3: Team Model — role, seniority, dedication %, phase, quantity (NO rates)
Sheet 4: Cost Driver Taxonomy — category, driver, magnitude, phase, owner
Sheet 5: Timeline — feature, optimistic, probable, pessimistic, PERT, critical path
Sheet 6: Magnitude Scenarios — optimistic, probable, pessimistic (FTE-months)
Sheet 7: Sensitivity Analysis — driver, impact on magnitude, risk level
Formato DOCX (bajo demanda):
06_Cost_Drivers_{TIPO_SERVICIO}_{project}_{WIP}.docx
Via python-docx con Design System MetodologIA v5. Cover page, TOC auto, headers/footers branded, tablas zebra. Poppins headings (navy), Montserrat body, gold accents.
Formato PPTX (bajo demanda):
{fase}_{entregable}_{cliente}_{WIP}.pptx| Dimension | Peso | Criterio |
|---|---|---|
| Trigger Accuracy | 10% | Activacion correcta ante keywords de cost estimation, effort drivers, sizing, team composition, PERT, Monte Carlo, Phase 4. |
| Completeness | 25% | 7 secciones cubren scope, sizing, team, drivers, timeline, magnitude, y governance. 8+ categorias de drivers. Zero precios en output. |
| Clarity | 20% | Magnitudes en FTE-meses (nunca dinero). Rangos con P50/P80/P95. Disclaimer de costeo presente y claro. |
| Robustness | 20% | Edge cases (precio final solicitado, greenfield, legacy, multi-vendor, regulatory, scope ambiguo) manejados. 8 service types soportados. |
| Efficiency | 10% | Variante ejecutiva reduce a S1+S4+S6 (~40%). Triangulacion flag automatico cuando divergencia >30%. |
| Value Density | 15% | Driver taxonomy accionable (owner por driver). Sensitivity analysis identifica drivers que mas mueven la aguja. Innovation margin 5% explicito. |
Umbral minimo: 7/10. Debajo de este umbral, revisar triangulacion de sizing y completeness de driver taxonomy.
Autor: Javier Montano · Comunidad MetodologIA | Ultima actualizacion: 15 de marzo de 2026