Idea to Plan: AI Skill for Product Management
Idea to Plan is an AI skill that walks a raw idea to a build-ready plan: problem, discovery, users, riskiest-assumption test, priorities, then spec. Evidence comes before the PRD, always.
npx skills add Uxcel-Lab/product-skills --skill pm-idea-to-plan
Copied
What this skill does
Stops the rush to a plausible-looking PRD and sequences the work an idea needs before anyone builds.
- Establishes the context once: idea, stakes, stage, market maturity, and evidence in hand.
- Frames the problem as an evidence-backed gap before anything else runs.
- Runs discovery before the spec, scaled to the stakes of the build decision.
- Grounds personas and jobs-to-be-done in the discovery evidence, not guesses.
- Foregrounds the riskiest-assumption test as an explicit build gate.
- Prioritizes the first slice against validated risk, not a wish list.
- Writes the spec last and keeps it proportionate to the evidence beneath it.
- Hands the finished plan to product-review for one prioritized report.
When to use it
Use it when you need an AI assistant to:
- go from a raw idea to a build-ready plan;
- start a new product without rushing to a PRD;
- sequence discovery, validation, and spec work in order;
- name and test the riskiest assumption before building;
- plan a product bet in a crowded market;
- hand a finished plan to a full review.
Skills this workflow composes
| Skill | Role |
|---|---|
| Problem statement | Frames the problem as a gap with evidence, first |
| Discovery | Establishes who adopts, whether they'd switch, and whether they'd pay |
| Personas & JTBD | Grounds the target users in the discovery evidence |
| Assumption testing | Designs the cheapest test of the make-or-break belief, as a build gate |
| Prioritization | Picks the first slice against validated risk, trade-offs exposed |
| Product spec | Writes the PRD last, proportionate to the evidence beneath it |
| OKRs & KPIs | Offered when the plan needs first-class outcome metrics |
How the skill works
1
Establish the context
The skill pins the idea, stakes, stage, market maturity, and the evidence already in hand.
2
Climb the ladder
It composes problem, discovery, users, assumption test, prioritization, and spec in order, gating each rung.
3
Validate the plan
It hands the finished plan to product-review for one merged, severity-rated report.
