How to use the Prioritize Feature Backlog agent skill

What is an agent skill, and what does this one do?

An agent skill is a folder containing a SKILL.md file. The file holds instructions that an AI assistant reads when a task matches the skill's description. This skill's description covers prioritizing a feature backlog, making scope decisions and ranking product ideas.

When you give the assistant a list of feature ideas, the skill tells it to score each one on four things: impact, effort, risk and strategic alignment with your product goals. It then recommends a ranked top 5. For each pick it explains the rationale, the trade-offs considered, and what was deprioritized and why. It can also present the results as a prioritization table.

• Impact: how much the feature moves the outcomes you care about, using Opportunity Score if customer data is available.
• Effort: development, design and coordination required.
• Risk: how much uncertainty exists and which assumptions need testing.
• Strategic alignment: how well the feature fits the product vision and current goals.

Install the prioritize-features skill

Download the ZIP file from the skill's page on this site. Unzip it to get the prioritize-features folder containing SKILL.md. Put that folder in one of these locations:

To make the skill available across your projects, unzip it into ~/.claude/skills/. To make it available only inside one project, unzip it into that project's .claude/skills/ folder.

You can also upload the ZIP in the Claude apps under Settings → Capabilities → Skills.

• Personal install: ~/.claude/skills/
• Project install: .claude/skills/ inside your project folder
• Claude apps: Settings → Capabilities → Skills, then upload the ZIP

Trigger the skill with natural requests

You do not need a special command. The skill is meant to be picked up when your request matches its description, so ask in ordinary words. Phrases that match the skill's stated use cases include prioritizing a feature backlog, making scope decisions, and ranking product ideas.

The skill works best when you state your product objective and success metrics up front, because the instructions begin by confirming them. If you have a spreadsheet or backlog file, attach it or point the assistant to it, since the skill tells the assistant to read and analyze provided files directly.

• "Prioritize this feature backlog for our Q3 goal of reducing churn. Give me the top 5 with reasons."
• "We have a release with room for five features. Help me make the scope decision from this list."
• "Rank these product ideas and show the trade-offs in a table."
• "Here is our backlog spreadsheet. Score each item and suggest whether we should use ICE or RICE."

What the output looks like

The skill asks for a clear ranking from 1 to 5, a brief rationale for each selection, the key trade-offs, and a list of what was deprioritized with reasons. It may also produce a prioritization table. If the output is substantial, the instructions tell the assistant to save it as markdown.

The skill's framework guidance recommends Opportunity Score for evaluating customer problems, using Importance multiplied by (1 minus Satisfaction), normalized to 0 to 1. It also describes ICE, where Impact is Opportunity Score times number of customers, multiplied by Confidence and Ease, and RICE, which adds Reach as a separate factor for larger teams. The skill advises prioritizing problems (opportunities) rather than solutions.

What you need and safety notes

The skill lists no required tools or packages. It does not run scripts and does not use the network, according to its listing. It is licensed under MIT.

The skill is a lightweight framework guide. Its scores are only as good as the inputs. If you have no real customer data, the assistant is working from guesses, and the resulting numbers can look more precise than they are. Treat the ranking as a structured starting point for discussion, not as evidence.

• Requires: nothing listed
• Runs scripts: no
• Uses network: no
• Licence: MIT

Tips for better results

Give the assistant real inputs. Customer data, usage numbers and team capacity make the impact and effort scores more grounded. Ask it to name the assumptions behind each score, since the skill's risk criterion is about what needs testing.

Ask for the deprioritized list as well as the top 5. The trade-off explanations are often the most useful part when you need to justify a decision to stakeholders.

• Name the product objective and success metric in your first message.
• Attach the backlog file rather than pasting a long list.
• Ask which scoring method fits your team size and data, and why.
• Re-run with different assumptions to see whether the top 5 changes.