SKILL FILE

Prioritization Engine with AI

Defend your roadmap with RICE / ICE scoring frameworks. Show stakeholders exactly WHY features ranked the way they did — with traceable scoring + sensitivity analysis.

Download Skill File ↓

What this skill file teaches Claude

Drop one markdown file into your repo. Claude Code learns how to run this entire workflow.

1

RICE scoring per item

Reach × Impact × Confidence ÷ Effort — calculated and ranked across your backlog.

2

Scoring rationale

Each score includes the reasoning + source. No mystery numbers.

3

Value/Effort 2×2

Visual quadrant: Do Now / Do Next / Quick Wins / Defer / Deprioritize.

4

Sensitivity analysis

Shows what would have to change for the ranking to flip. Tests how robust the prioritization is.

5

Recommendation buckets

Explicit recommendations per item — "do now", "defer", "kill" — with reasoning.

6

Assumption flagging

Surfaces which scores rely on unvalidated assumptions, so you know what to test.

What you can build with this

Rank 3+ competing initiatives

When the roadmap has more contenders than slots, the skill turns vibes into evidence.

Defend prioritization to stakeholders

Show the math. Disagree with the numbers if you want — but the conversation is now grounded.

Justify "no" data-driven

A defensible "no" is more useful than a vague "later." The skill produces the no with reasoning.

Get the full skill file

Everything above is 80% of the skill file. Download the complete version with full implementation details, agent prompts, and ready-to-run scripts.

Common questions

Reach × Impact × Confidence ÷ Effort. A scoring framework that combines how many users a feature would affect, how much it'd help them, how confident you are in those estimates, and how much it costs to build.
ICE (Impact × Confidence × Ease) is simpler and works for smaller comparisons (under 10 items). RICE is more rigorous and scales to bigger backlogs. The skill supports both.
The skill prompts you for the data source per Reach estimate — analytics, support volume, survey responses. If you don't have any, it flags the score as low-confidence.
That's the point — the scoring becomes the conversation. Disagree on Impact? Talk about which user segment matters more. The framework forces the debate to be about substance, not vibes.
Before every batch you ship. If you're shipping weekly, re-score weekly. After every learning event (a customer churns, a competitor launches, a big piece of feedback lands). Quarterly is the cadence big companies pretend to follow; for a solo founder it's whenever the data actually changes, which is usually faster than you think.

Ready to automate?

SoloStack gives you every skill pre-installed — scraping, marketing, sales, CRM, and more. One repo. Every department.

Book a Call →