prioritize-features
by phurynprioritize-features skill for Product Management helps rank a feature backlog into a defensible top 5 using impact, effort, risk, and strategic fit. Use it when you need to compare feature ideas, make scope decisions, and explain why one item should come first.
This skill scores 78/100, which means it is a solid listing candidate for directory users: it has a clear trigger, a real prioritization workflow, and enough guidance to be more useful than a generic prompt, though it still leaves some execution details to the agent.
- Clear, specific trigger for prioritizing feature backlogs and ranking product ideas.
- Gives an explicit workflow: confirm objective, evaluate features, and rank the top 5 recommendations.
- Includes framework guidance (Opportunity Score, ICE, RICE) that helps agents choose an appropriate prioritization method.
- The excerpt shows no scripts, references, or support files, so there is limited external validation or tooling support.
- Some instruction text is truncated in the evidence, which may leave edge-case handling and execution details less explicit than ideal.
Overview of prioritize-features skill
The prioritize-features skill helps you rank a feature backlog into a defensible top 5 based on impact, effort, risk, and strategic fit. It is best for Product Management work where you need to turn a long list of ideas into a clear recommendation, not just brainstorm more options.
Use this prioritize-features skill when you have feature candidates, a product goal, or competing stakeholder requests and need a decision you can explain. It is especially useful if you want a structured output that goes beyond a generic prompt and forces tradeoff thinking.
Best fit for Product Management
prioritize-features for Product Management is a strong fit when your job is to choose what to build next, align teams, or prepare for roadmap discussion. It works best when you already have a backlog, rough customer evidence, or at least a stated business objective.
What it actually optimizes
The workflow is designed to evaluate each idea against impact, effort, risk, and strategic alignment, then recommend the top 5. The practical value is not a perfect scorecard; it is a repeatable way to make prioritization less subjective and easier to defend.
When it is most useful
Choose this skill when the question is “what should we do first?” rather than “what should we build from scratch?” If you need opportunity sizing, backlog ordering, or a concise recommendation for leadership, this skill gives you a better starting point than an open-ended product prompt.
How to Use prioritize-features skill
Install the skill
Use the repository install flow for a targeted skill add, for example: npx skills add phuryn/pm-skills --skill prioritize-features. If your environment uses a different skills manager, install the prioritize-features skill into the same workspace where the product context files live so the agent can read them.
Give it decision-ready input
The skill works best when you provide three things: the product objective, the feature list, and any constraints. A weak request is “prioritize these ideas.” A stronger request is “prioritize these 12 ideas for activation in Q3, optimizing for trial-to-paid conversion, with one designer and two engineers available.”
A prompt shape that gets better output
For prioritize-features usage, include:
- the target user or segment
- the outcome you want to move
- the feature ideas in a plain list
- any data you have, such as usage, customer requests, or churn reasons
- constraints like timeline, team capacity, platform limits, or dependencies
Example:
“Prioritize these feature ideas for a B2B admin product. Goal: reduce onboarding drop-off by 15%. Team: 2 engineers, 1 designer, 6 weeks. Use the feature list below and return the top 5 with a short rationale and key risks.”
Read these files first
Start with SKILL.md because it defines the workflow, evaluation logic, and what the skill expects from $ARGUMENTS. If you need broader context, also inspect any sibling skill or taxonomy files in the repo, especially prioritization-frameworks, so you can match the right scoring model to the decision.
prioritize-features skill FAQ
Is this just a better prompt?
No. The prioritize-features skill is useful because it encodes a repeatable prioritization workflow and pushes the agent to compare ideas on the same criteria. A normal prompt may produce a list; this skill is meant to produce a decision.
Do I need customer data?
No, but the result is stronger when you have it. If you can provide usage data, customer feedback, or opportunity signals, the skill can make the prioritization more grounded and less opinion-based.
When should I not use it?
Do not use it if you want ideation, not ranking, or if the feature list is too vague to compare. If your real problem is “what are possible solutions?” or “what problem should we solve?”, you may get better results by framing the work around discovery first.
Is it beginner-friendly?
Yes, if you can describe a product goal and a list of candidate features. You do not need advanced scoring knowledge to use the prioritize-features skill; the main requirement is providing enough context for tradeoff decisions.
How to Improve prioritize-features skill
Give stronger constraints
The best outputs come from real limits, not broad aspirations. Add timeline, staffing, platform, and launch constraints so the skill can separate “valuable” from “valuable and feasible now.”
Distinguish problems from solutions
A common failure mode is handing the agent a list of proposed solutions without the underlying problem statement. For better prioritize-features guide results, describe the user problem or business objective first, then attach the feature ideas underneath it.
Add evidence, not just opinions
If you have it, include customer interviews, support themes, funnel drop-off points, revenue impact, or churn reasons. That lets the skill weigh confidence and importance more intelligently instead of treating every idea as equally plausible.
Iterate after the first pass
Use the first ranking to expose disagreements: ask which items are blocked by dependencies, which are high effort for low strategic value, and which would move if the target metric changed. That second pass usually improves the recommendation more than expanding the original list.