Prioritization Rigor Audit: AI Skill for Product Management
Prioritization Rigor Audit is an AI skill that traces every score and ranking to its evidence and strategy. It flags invented numbers and silent trade-offs, with a severity rating and concrete fix per finding.
npx skills add Uxcel-Lab/product-skills --skill pm-prioritization-rigor-audit
What this skill does
Checks whether a scored backlog or roadmap order would survive the question: why this, why now?
- Traces every RICE or ICE score to its evidence and flags decoration.
- Checks rankings trace to strategy instead of loudness, recency, or one big deal.
- Flags feature-factory lists that name outputs instead of problems and outcomes.
- Challenges must-have inflation and demands an explicit won't-have list.
- Surfaces silent trade-offs: what was rejected, what the choice costs, what would change it.
- Tests the framework fits the data and the stakes of the decision.
- Scales rigor to reversibility, so quick calls on small items stay quick.
- Rates findings by decision damage and ends with the top three fixes.
When to use it
Use it when you need an AI assistant to:
- audit a scored backlog before locking the quarter;
- check a RICE, ICE, or MoSCoW exercise for false precision;
- trace a roadmap order back to the product strategy;
- deflate an everything-is-critical must-have list;
- make the trade-offs behind a ranking visible;
- validate any "what should we build first" decision.
What this audit checks
| Category | What it catches |
|---|---|
| Score evidence | Invented RICE inputs, false precision, scoring applied after the decision |
| Strategy linkage | Rankings ordered by loudness, recency, or a single deal |
| Outcome orientation | Feature-name lists with no problems or affected-user counts |
| Honest categories | Must-have inflation, empty won't-have lists, inverted Kano logic |
| Explicit trade-offs | No record of what was rejected or what the choice costs |
| Framework fit | RICE with no data, guessed Kano labels, framework-hopping |
How the skill works
1
Establish stakes
The skill confirms decision stakes, available data, deadline pressure, and the framework in use.
2
Trace every ranking
It traces scores to evidence, rankings to strategy, and checks categories and trade-offs for honesty.
3
Deliver ranked findings
It rates each finding by decision damage and lists the top three fixes with questions to answer.
Try it with
Related skills
1.0.0
MIT
Product Management
Audits
