メニュー
Use when defining product direction, vision document sections, or roadmap. Triggers on strategic product planning, vision creation, or vision document quality review.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
Use when producing product artifacts - epics, user stories, acceptance criteria, or analytics specs.
Use when designing UI, choosing colors, typography, spacing, creating visual hierarchy, or making design decisions. Based on Refactoring UI by Wathan & Schoger.
Use when someone needs a plain-language explanation of how an existing feature or edge case behaves in the codebase. Triggers - non-technical stakeholder asks how something works, need code-informed functional explanation, need to understand rules or edge cases without implementation planning.
Use when a developer wants to go from rough idea to working code autonomously. Triggers - imperfect prompt, solo dev wants full pipeline, "just build it", idea to implementation without manual steps.
Use when creating a new agentic workflow skill. Triggers - need new multi-step orchestrated process, want to add workflow to agentic framework, building spec-driven pipeline.
Use when building frontend features from idea to implementation. Triggers on new UI components, pages, or features needing design and code.
SOC 職業分類に基づく
| name | agentic:skill:product-vision |
| description | Use when defining product direction, vision document sections, or roadmap. Triggers on strategic product planning, vision creation, or vision document quality review. |
Guide for writing comprehensive product vision documents. Each section has quality criteria, frameworks, and anti-patterns.
.claude/_agentic_output/product/vision/{timestamp}-{main-topic}.md
Topic:
Timestamp: (ISO)
Status: Draft | Active | Superseded
Owner: CPO
Framework: Geoffrey Moore's positioning — "For [target user] who [need], [product] is a [category] that [key benefit]. Unlike [alternative], we [differentiator]."
Quality test: A stranger reads it in 60 seconds, can repeat back what it does, who it's for, and why it wins.
Anti-patterns: Feature lists, jargon, burying the value proposition.
Framework: "From X to Y" — describe current state → desired future state. One sentence, present tense, customer-centric.
Quality test: Is it memorable and repeatable? Would it inspire someone to join the team? Time horizon 3-10 years.
Anti-patterns: Too vague ("make the world better"), too narrow (describes today's product), confusing vision with mission.
Frameworks: Five Whys (symptom → root cause), JTBD ("When [situation], I want to [motivation], so I can [outcome]").
Must include: Symptoms vs root causes (separated explicitly), quantified impact (cost, time, churn), emotional/social dimensions.
Quality test: Reader can feel the pain, see the evidence, understand why solving this is worth investing in.
Anti-patterns: Describing problems in terms of your solution ("users need a dashboard"), surface-level symptoms without root causes, no evidence.
Frameworks: JTBD + Personas combined (personas = who, JTBD = what they're trying to accomplish), behavioral segmentation over demographics.
Per persona include: Role/context, goals, pain points, current workarounds, success criteria.
Must include: Anti-personas (who you're NOT building for). Max 3-4 primary personas.
Quality test: Each persona has a clear JTBD, measurable pain, and you can point to real users who match.
Anti-patterns: "Persona theater" (beautiful docs nobody uses), too many personas, demographics-heavy behavior-light.
Frameworks: SWOT (internal strengths/weaknesses vs external opportunities/threats), Porter's Five Forces, Blue Ocean Strategy Canvas.
Must include: Competitive feature matrix, market sizing (TAM/SAM/SOM with methodology), macro trends (tailwinds/headwinds), honest about threats.
Quality test: Reader understands why now, who the real competition is, where whitespace lives.
Anti-patterns: Cherry-picking data, ignoring indirect competitors, stale data, confusing large TAM with real opportunity.
Frameworks: Value Proposition Canvas (customer pains/gains ↔ pain relievers/gain creators), 10x Test (is solution 10x better on at least one dimension?).
Must include: Direct line from each customer pain → specific product capability, defensible moats (network effects, data, switching costs, brand).
Quality test: Can explain why competitors can't easily replicate each differentiator.
Anti-patterns: Listing features instead of value, differentiators that competitors copy overnight, "we do everything" syndrome.
Format: "Even over" — "Simplicity even over feature completeness". 5-7 principles max.
Quality test for EACH principle: Would saying the opposite also be reasonable? If not, it's a platitude, not a principle. Can you find a real past decision it would have changed?
Anti-patterns: Generic truisms ("we care about quality"), too many (>7), written once never referenced.
Framework: Three Horizons Model — H1 (core, 0-6mo), H2 (emerging, 6-18mo), H3 (future bets, 18mo+). Goals cascade: short feeds medium feeds long.
Must include: Both product AND business goals per horizon. Different goal types: short = tactical wins, medium = capability building, long = market position.
Quality test: Can trace a line from "what we're building this quarter" to "why we'll win in 5 years."
Anti-patterns: Only short-term goals, long-term too vague, no connection between horizons.
Frameworks: OKRs (Objective = qualitative aspirational, Key Result = quantitative measurable), HEART (Happiness, Engagement, Adoption, Retention, Task success), Pirate Metrics (AARRR).
Must include: Baselines, targets, timeframes, measurement method. Mix of leading and lagging indicators. Max 3-5 objectives.
Quality test: Every metric connects to a strategic goal. Can explain WHY each metric matters.
Anti-patterns: Vanity metrics, too many OKRs, key results that are task lists, no baselines.
Framework: MoSCoW (Must/Should/Could/Won't). Must-Haves ~60% of scope.
Quality test: Each feature traces to a user need + strategic goal. Described as capabilities (what users can do), not implementation details.
Anti-patterns: Everything is "Must-Have", features disconnected from problem/personas, describing solutions instead of capabilities, missing Won't-Have.
Framework: NN/g Journey Mapping — actor, scenario, phases, actions, thoughts, emotions, opportunities.
Must include: Full experience (awareness → onboarding → core usage → expansion → advocacy), friction points, "moments of truth", both happy and error paths. 2-3 critical journeys max.
Quality test: Each journey reveals at least one actionable improvement opportunity.
Anti-patterns: Too granular or too abstract, missing emotional layer, only happy paths, never validated with users.
Frameworks: Business Model Canvas (9 blocks), SaaS metrics (ARR, churn, LTV/CAC ratio ≥3:1), value-based pricing.
Must include: How product makes money, unit economics, pricing model alignment with value perception, model evolution over time.
Quality test: Reader understands economics, whether they work, how model evolves.
Anti-patterns: "Figure out monetization later", copying competitor pricing blindly, ignoring unit economics.
Frameworks: Pre-mortem ("imagine it failed — why?"), Assumption Mapping (importance × uncertainty), Risk Matrix (likelihood × impact).
Must include: Separate risks (things that might happen) from assumptions (things believed true but unvalidated). Each risk: likelihood, impact, mitigation, owner. Each assumption: confidence level, validation plan.
Quality test: Team has thought deeply about what could go wrong and has a plan.
Anti-patterns: Checkbox exercise, only technical risks, assumptions listed but never validated, missing "assumption of demand".
Framework: Now / Next / Later — decreasing specificity over time. Theme-based, not feature-list.
Must include: Outcome-oriented themes per phase, entry/exit criteria, what's NOT in each phase, connected to strategic goals.
Quality test: Tells a story of value delivery over time, not just a Gantt chart.
Anti-patterns: Date-driven feature lists, equal specificity across all phases (false precision), no learning loops, treated as commitment.
Must include: Each question has owner, approach to answer, deadline. Categorized (technical, market, user, business). Prioritized by decision impact.
Quality test: Section shrinks over time. No question that should have been resolved before writing the document.
Anti-patterns: Empty (false confidence), vague questions, no resolution plan, never updated.