L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

feature-prioritizer

feature-prioritizer

Score a feature list using RICE (Reach, Impact, Confidence, Effort) and return a ranked table with recommendation. Use when the user wants to prioritize features, decide what to build next, or cut scope.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: feature-prioritizer description: Score a feature list using RICE (Reach, Impact, Confidence, Effort) and return a ranked table with recommendation. Use when the user wants to prioritize features, decide what to build next, or cut scope.

Ask the user:

  1. "Paste your list of features or initiatives."
  2. "What's the timeframe — one quarter, one sprint, one year?"
  3. "Any features that are already committed or non-negotiable? I'll mark them separately."

For each feature, estimate:

  • Reach — how many users affected per period (1–10 scale or absolute number)
  • Impact — how much does it move the needle per user (0.25 = minimal, 0.5 = low, 1 = medium, 2 = high, 3 = massive)
  • Confidence — how sure are you about Reach and Impact? (percentage: 50%, 80%, 100%)
  • Effort — person-months to build (use your best estimate; flag if unclear)

RICE score = (Reach × Impact × Confidence) / Effort


Feature Prioritization — [Product / Quarter]

FeatureReachImpactConfidenceEffortRICE ScorePriority
[feature][#][#][%][months][score]HIGH/MED/LOW

Recommended build order

Top 3 features with one-line rationale for each.

Features to cut or defer

Which items scored low and why you should drop or delay them.

What the scores can't tell you

Flag 1–2 strategic or political factors that aren't captured in RICE but matter for the decision.


State your assumptions clearly. Ask: "Want me to re-score with different estimates?"

Skills enregistrées