Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply prioritisation frameworks (RICE, MoSCoW, Kano, ICE, Opportunity Scoring) to rank features and backlog items. Use when asked to prioritise features, rank a backlog, decide what to build next, or evaluate tradeoffs between competing ideas. Produces a scored, ranked feature list with framework-specific tables, recommended build order, deprioritised items, and assumptions made.
.claude/skills/mohitagw15856-feature-prioritisation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 35% | 0% |
Aplicar el marco de priorización correcto a cualquier backlog y producir una clasificación clara y defendible con justificación — no solo una lista ordenada.
Solicitar al usuario estos datos si no se proporcionan:
Preguntar al usuario qué marco prefiere, o recomendar basado en contexto:
| Situación | Marco Recomendado | |---|---| | Necesitar una puntuación rápida basada en datos | RICE | | Reunión de alineación de stakeholders | MoSCoW | | Entender deleite del cliente vs expectativas | Kano | | Startup en fase temprana, decisiones rápidas | ICE | | Identificar necesidades de clientes infraservidas | Opportunity Scoring | | Decisiones estratégicas de cartera | Matriz de Valor vs Esfuerzo |
Fórmula: (Alcance × Impacto × Confianza) ÷ Esfuerzo
| Factor | Definición | Escala | |---|---|---| | Alcance | Usuarios impactados por trimestre | Número real | | Impacto | Efecto en el objetivo por usuario | 0.25 / 0.5 / 1 / 2 / 3 | | Confianza | Qué tan seguro estás? | 50% / 80% / 100% | | Esfuerzo | Personas-mes requeridos | Número real |
Tabla de salida: | Característica | Alcance | Impacto | Confianza | Esfuerzo | Puntuación RICE | Prioridad | |---|---|---|---|---|---|---|
Categorizar cada característica como:
Siempre preguntar: "Must have para qué?" — definir el alcance (lanzamiento, sprint, trimestre) antes de categorizar.
Fórmula: Impacto + Confianza + Facilidad (cada una 1–10)
Rápido, subjetivo — bueno para decisiones tempranas antes de que existan datos.
Clasificar características en:
Recomendar construir: todas las características Basic primero → características Performance para casos de uso clave → 1–2 características Excitement por release.
Este skill incluye un script Python que solo usa stdlib y que calcula la clasificación para los marcos basados en matemáticas (RICE, ICE) para que la puntuación de características sea consistente entre sesiones.
bash# RICE desde JSON python3 scripts/feature_prioritisation.py initiatives.json --framework rice # RICE desde CSV python3 scripts/feature_prioritisation.py initiatives.csv --framework rice --format csv # ICE desde JSON python3 scripts/feature_prioritisation.py features.json --framework ice # Pasar mediante pipe printf '%s\n' '[{"name":"API refactor","impact":8,"confidence":80,"ease":5}]' \ | python3 scripts/feature_prioritisation.py --framework ice -
Usar --json para producir salida legible por máquina para herramientas posteriores.
Marco Utilizado: RICE / MoSCoW / ICE / Kano / Personalizado] Alcance: Sprint / Trimestre / Release] Objetivo siendo priorizado: Métrica u objetivo]
Tabla puntuada usando marco seleccionado]
Orden de Construcción Recomendado:
Explícitamente Depriorizadas:
Supuestos Realizados:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 12,948 | 10,488 | -19% | 1 | 1 | 0% | 2,884 | 3,864 | +34% | 0 | 0 | — |
case-02 | pass→pass | 9,251 | 9,963 | +8% | 1 | 1 | 0% | 1,969 | 3,615 | +84% | 0 | 0 | — |
case-03 | pass→fail | 13,485 | 9,220 | -32% | 1 | 1 | 0% | 2,513 | 3,381 | +35% | 0 | 0 | — |
case-04 | pass→pass | 12,657 | 11,392 | -10% | 1 | 1 | 0% | 2,248 | 3,548 | +58% | 0 | 0 | — |
case-05 | fail→fail | 9,186 | 7,894 | -14% | 1 | 1 | 0% | 1,740 | 2,950 | +70% | 0 | 0 | — |
case-06 | pass→pass | 9,364 | 10,324 | +10% | 1 | 1 | 0% | 1,695 | 3,215 | +90% | 0 | 0 | — |
case-07 | fail→pass | 13,504 | 9,165 | -32% | 1 | 1 | 0% | 2,431 | 3,194 | +31% | 0 | 0 | — |
case-08 | pass→pass | 11,915 | 13,174 | +11% | 1 | 1 | 0% | 2,221 | 3,919 | +76% | 0 | 0 | — |
case-09 | pass→pass | 8,114 | 6,821 | -16% | 1 | 1 | 0% | 1,593 | 2,886 | +81% | 0 | 0 | — |
case-10 | fail→pass | 12,033 | 8,916 | -26% | 1 | 1 | 0% | 2,279 | 3,079 | +35% | 0 | 0 | — |
case-11 | pass→pass | 7,523 | 5,237 | -30% | 1 | 1 | 0% | 1,406 | 2,516 | +79% | 0 | 0 | — |
case-12 | pass→pass | 7,370 | 2,368 | -68% | 1 | 1 | 0% | 1,521 | 2,024 | +33% | 0 | 0 | — |
case-13 | pass→pass | 5,782 | 1,512 | -74% | 1 | 1 | 0% | 1,143 | 1,795 | +57% | 0 | 0 | — |
case-14 | fail→pass | 11,145 | 10,061 | -10% | 1 | 1 | 0% | 2,020 | 3,326 | +65% | 0 | 0 | — |
case-15 | pass→pass | 11,201 | 11,000 | -2% | 1 | 1 | 0% | 2,069 | 3,343 | +62% | 0 | 0 | — |
case-16 | pass→pass | 11,521 | 13,839 | +20% | 1 | 1 | 0% | 1,888 | 3,840 | +103% | 0 | 0 | — |
case-17 | fail→pass | 11,165 | 7,661 | -31% | 1 | 1 | 0% | 2,012 | 2,811 | +40% | 0 | 0 | — |
case-18 | fail→fail | 12,812 | 10,158 | -21% | 1 | 1 | 0% | 2,227 | 3,254 | +46% | 0 | 0 | — |
case-19 | fail→fail | 12,896 | 7,342 | -43% | 1 | 1 | 0% | 2,827 | 3,083 | +9% | 0 | 0 | — |
case-20 | fail→fail | 16,684 | 16,703 | +0% | 1 | 1 | 0% | 3,334 | 4,643 | +39% | 0 | 0 | — |
case-21 | fail→fail | 19,015 | 16,506 | -13% | 1 | 1 | 0% | 3,873 | 4,732 | +22% | 0 | 0 | — |
case-22 | fail→fail | 26,652 | 25,907 | -3% | 1 | 1 | 0% | 6,179 | 7,683 | +24% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.