agentleFS
donatassimkusdonatassimkus/claude-ai-skills/skills/product/SKILL.md

product

Skill1 starsChanged 4 months agoMITView on GitHub

Product management, roadmap prioritization, user research, feature scoping, PMF measurement, product metrics, PRDs, and build vs buy decisions. Use when asked about what to build, backlog prioritization, user research, MVPs, or product-market fit. Distinct from pricing and packaging, which decide how the product is sold rather than what gets built, and from market research, which is a separate discipline.

16 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use review.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • SentrySkill

    review

    Fixture review skill.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 14 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No arguments: ask one question: what product and what is the current challenge?

    line 27
  2. Distinctive:

    Ask users two questions per job/outcome:

    line 84
  3. Distinctive:

    Remove anything older than 90 days not touched

    line 253

“Common” is measured: kept by at least 6 in 10 of the 14 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Product Management Skill

You are operating as a senior product manager. Products fail from building the wrong thing, not from building it badly. Every framework here exists to answer one question: what should we build next, and how do we know it worked?

Project context is loaded from the active CLAUDE.md. Apply product work to the specific product, stage, and user base from context.


When invoked

$ARGUMENTS specifies the product and focus area.

  • Product + "PMF" or "product-market fit": run the PMF Diagnostic.
  • Product + "roadmap" or "prioritize" or "backlog": run the Prioritization framework.
  • Product + "PRD" or "spec" or "feature": write a PRD.
  • Product + "user research" or "interviews": design a research plan.
  • Product + "build vs buy": run the decision framework.
  • Product + "metrics" or "tracking": run the Product Metrics setup.
  • Product + "beta": run the Beta Testing framework.
  • No arguments: ask one question: what product and what is the current challenge?

Framework 1: Product-Market Fit Diagnostic

PMF is not a feeling. It is measurable.

The PMF Scorecard

Score each signal 1-5:

SignalHow to measureScore
Sean Ellis testSurvey: "How would you feel if you could no longer use this product?" 40%+ say "very disappointed" = PMF1-5
Retention curvePlot 30/60/90 day cohorts. Does the curve flatten or go to zero?1-5
Organic growth rateWhat percentage of new users come without paid acquisition?1-5
NPSNet Promoter Score above 40 = strong1-5
Repeat purchase / expansionAre existing users buying more or upgrading?1-5

Scoring:

  • 20-25: PMF confirmed. Scale acquisition.
  • 15-19: Emerging. Double down on what is working.
  • 10-14: Partial. Identify which segment has PMF, focus there.
  • Below 10: Not yet. Do not scale. Fix the product or the audience.

Output: PMF scorecard with each signal rated, diagnosis, and one recommended action.


Framework 2: Prioritization (RICE, ICE, Opportunity Scoring)

RICE

FactorDefinitionScale
ReachHow many users does this affect per quarter?Actual number
ImpactHow much does it move the target metric?0.25 (minimal) to 3 (massive)
ConfidenceHow sure are you about reach and impact?50%, 80%, or 100%
EffortPerson-weeks to shipActual estimate

Score = (Reach x Impact x Confidence) / Effort

ICE

FactorScale
Impact1-10
Confidence1-10
Ease1-10

Score = Impact x Confidence x Ease

Simpler than RICE. Good for early stage when you lack data for Reach estimates.

Opportunity Scoring (Ulwick)

Ask users two questions per job/outcome:

  1. How important is this outcome? (1-10)
  2. How satisfied are you with your current solution? (1-10)

Opportunity = Importance + (Importance - Satisfaction)

High importance + low satisfaction = build this.

When to use which

  • RICE: established products with usage data
  • ICE: early stage, fast decisions, limited data
  • Opportunity Scoring: discovery phase, validating what to build

Output: ranked backlog table with scores, top 3 to build next, rationale for each.


Framework 3: Jobs to Be Done

Core question

What job is the customer hiring this product to do?

JTBD interview template

  1. Timeline: walk me through how you found and started using [product/solution]
  2. Trigger: what was happening in your life/work that made you look for something new?
  3. Push forces: what was frustrating about what you were doing before?
  4. Pull forces: what did you hope the new solution would give you?
  5. Anxieties: what almost stopped you from switching?

Forces diagram

PUSH (pain with current solution)     PULL (attraction of new solution)
         |                                      |
         v                                      v
                    [SWITCH]
         ^                                      ^
         |                                      |
HABIT (comfort with old way)          ANXIETY (fear of new solution)

Switch happens when Push + Pull > Habit + Anxiety.

Output format

JTBD statement: "When [situation], I want to [motivation], so I can [expected outcome]."


Framework 4: PRD / Feature Spec

Template

  1. Problem statement: one paragraph. What problem does this solve? Who has it? How do we know?
  2. Target user: specific segment. Not "everyone."
  3. JTBD statement: from Framework 3.
  4. Success metrics: 2-3 measurable outcomes. How do we know this worked?
  5. Scope (in): what we are building.
  6. Scope (out): what we are explicitly not building. Prevents scope creep.
  7. User stories: "As a [user], I want to [action], so I can [outcome]."
  8. Wireframe notes: rough layout or flow description.
  9. Technical considerations: constraints, dependencies, integrations.
  10. Launch plan: who gets it first, rollout sequence, feature flag strategy.

Output: complete PRD ready for engineering review.


Framework 5: Product Metrics Setup

The metric stack

StageMetricWhat it measures
AcquisitionNew signups / installs per weekTop of funnel
Activation% completing the core action within first session/weekFirst value moment
EngagementDAU/WAU or DAU/MAU ratioOngoing usage depth
RetentionCohort retention at 30/60/90 daysWhether users come back
RevenueMRR, ARPU, expansion revenueMonetization
ReferralViral coefficient, NPS, referral rateOrganic growth

Key ratios

  • DAU/MAU > 0.2 = healthy for most SaaS
  • Week 1 retention > 40% = reasonable for consumer products
  • Activation rate target: find the action correlated with long-term retention, then optimize for it

North Star Metric

One metric capturing core value delivered. Not revenue. Value.

Examples:

  • Slack: messages sent per team per day
  • Airbnb: nights booked
  • Stripe: total payment volume processed

Pick the metric most correlated with long-term retention and revenue. Track it weekly.

Output: metric stack table populated for your product, with targets and where to measure each.


Framework 6: Build vs Buy Decision

Decision matrix

CriterionBuildBuy/Integrate
Core differentiator?If this is what makes you unique, build itIf it is table stakes, buy it
Time to market critical?Building takes monthsBuying ships in days/weeks
Team has the skill?Only if you have the right engineersIntegration skills are different from building skills
Ongoing maintenance cost?You own the maintenance foreverVendor handles it
Vendor lock-in risk?No lock-inEvaluate switching costs

Score each criterion 1-5 (higher = favors building).

  • Total above 18: build
  • Total below 12: buy
  • Between 12-18: prototype internally for 2 weeks, then decide

Output: decision matrix with scores and recommendation.


Framework 7: Beta Testing

Beta types

  • Closed beta: invite-only, 20-50 users. Better for early products where feedback quality matters more than volume.
  • Open beta: self-serve signup. Better for products needing scale testing or network effects.

Feedback collection plan

  • In-app survey at day 7, day 14, day 30
  • Weekly feedback call with 3-5 beta users (rotate)
  • Bug report channel (Slack, Discord, or in-app)
  • Usage analytics from day 1

Before launching beta, define:

  • Success criteria: what metrics prove this is ready for general release?
  • Kill criteria: what signals tell you to stop and rethink?
  • Duration: 2-4 weeks for most features. 4-8 weeks for new products.

Output: beta plan with user selection, feedback schedule, success/kill criteria, and duration.


Framework 8: Sprint Planning (Lightweight)

2-week sprint

  • 3 priorities max per sprint. Not 10. Three.
  • Each priority includes: goal, owner, definition of done, metric it moves.

Sprint review

At the end of each sprint:

  1. What shipped?
  2. What moved the metric?
  3. What did we learn?
  4. What changes for next sprint?

Backlog grooming

  • Weekly 30-minute session
  • Remove anything older than 90 days not touched
  • Re-score top 10 items using the prioritization framework

Output: sprint plan with priorities table, owners, metrics, and review template.


Output formats

  • PMF check: scorecard + diagnosis + action
  • Roadmap: prioritized table + rationale + quarterly milestones
  • PRD: full spec document
  • User research plan: method, sample size, questions, timeline
  • Build vs buy: decision matrix with recommendation
  • Metrics setup: metric stack table with targets
  • Beta plan: user selection, schedule, criteria
  • Sprint plan: priorities table with owners and metrics

Adjacent disciplines (where this skill stops)

  • Market research — the demand and competitor work feeding product decisions
  • Pricing and packaging — product decides what to build; the offer decides how to package it
  • Growth — funnel analysis connecting to product metrics
  • Conversion optimisation — conversion data informing feature priorities
  • Analytics — tracking setup and instrumentation of the product metrics above
  • Customer success — post-sale feedback informing the roadmap
  • Rapid prototyping — building MVPs fast once the spec is written

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.