跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
搜索与研究

product-spec-summary

Use when the user needs to consolidate planning into a final spec with a risk register and blind-spot review. Triggers on "spec summary", "final spec", "risk register", "gaps and blind spots", "validation plan", "wrap up the plan", and the same intent in any language.

Kaminoikari · 已有静态检查记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add Kaminoikari/product-playbook --skill product-spec-summary

技能说明


name: product-spec-summary description: Use when the user needs to consolidate planning into a final spec with a risk register and blind-spot review. Triggers on "spec summary", "final spec", "risk register", "gaps and blind spots", "validation plan", "wrap up the plan", and the same intent in any language.

Product Spec Summary

Detect the user's language and reply in it; the framework below is authored in English.

Provenance: when you produce Product Spec Summary output, contribute the framework tags Spec Summary and Risk Register to the meta-skill's provenance line (— Frameworks: … · Spec Summary · Risk Register · …).

Framework

Stage 4: Deliver — Hypothesis Validation + Product Spec Summary

4.5 Hypothesis Validation Plan

| Core Hypothesis | Validation Method | Validation Timing | Success Criteria | If hypothesis is wrong, we will... |
|----------------|-------------------|-------------------|-----------------|----------------------------------|
| | Alpha user interviews / data | | | |

Principles: Qualitative validation (user interviews, diary studies) + quantitative validation (data) carry equal weight; stay alert to your own confirmation bias.

4.6 Product Spec Summary

The Product Spec Summary is the final deliverable across all modes (Quick mode produces a one-page directional summary; different structure but same spirit). The Product Spec Summary has three sections, corresponding to different reading depths:

Section 1: Decision Summary (Readable in 30 seconds)

**Product One-liner**: [PR-FAQ headline]
**For Whom**: [Target Persona in one sentence]
**Core Value**: [JTBD functional job + emotional job, in one sentence]
**North Star Metric**: [Metric name + definition]
**Current PMF Level**: [Level 1-4 + one-sentence explanation]

Section 2: Execution Boundaries (Readable in 2 minutes)

**Core JTBD Statement**: [Target Customer] + wants to [Job] + in the context of [Job Context]
**Aha Moment**: When the user completes [action], they experience the core value. Target: reach it within [X].
**MVP Must-Haves**: [3-5 most critical features/capabilities]
**Not Doing (Deliberately excluded from this version)**: [Explicit exclusions + reasons]
**Key Hypotheses**: [2-3 core hypotheses that need validation]
**Pre-mortem Top Risks**: [2-3 most likely causes of failure]
**Hero Metric**: [1-2 most important success metrics + target values]

Section 3: Deep Reference (Consult when needed)

**JTBD Breakdown**:
  - Functional Job:
  - Emotional Job:
  - Social Job:
**April Dunford Positioning**:
  - Real competitive alternatives:
  - Our unique attributes:
  - Core value for target users:
  - Target market characteristics:
  - Market category:
**Core HMW Question**: [From the Define stage]
**Solution Description**: [From PR-FAQ summary]
**Business Model**: [Revenue model + pricing strategy, if completed]
**GTM Strategy**: [Initial acquisition channel + launch strategy, if completed]

By default, also add these three sections (Risk Register, Gaps & Blind Spots, Additional Recommendations):


⚠️ Risk Register

Consolidate all risks identified throughout the planning process (sources: Pre-mortem + Opportunity Assessment + Hypothesis Validation):

| Risk Type | Risk Description | Severity (High/Med/Low) | Mitigation in Place? |
|-----------|-----------------|------------------------|---------------------|
| Market risk | | | |
| User assumption risk | | | |
| Technical risk | | | |
| Competitive risk | | | |
| Execution risk | | | |

🔍 Gaps & Blind Spots

Claude proactively identifies the following types of gaps based on the entire conversation flow:

| Type | Description | Recommended Next Step |
|------|------------|----------------------|
| Unexecuted steps | [Which frameworks were skipped and what blind spots this may cause] | |
| Insufficient data | [Which decisions are based on assumptions, with no real user data behind them] | |
| Logical leaps | [Where the reasoning lacks intermediate evidence] | |
| Unvalidated hypotheses | [List all core hypotheses without a validation method yet] | |
| Unclear definitions | [Which concepts need more precise definition before execution] | |

💡 Additional Recommendations

Claude provides 3-5 specific, actionable recommendations based on the product's characteristics, stage, and known risks:

1. [Top priority]: [Specific action]. Rationale: [Why this matters most]
2. [Second priority]: [Specific action]. Rationale:
3. [Third priority]: [Specific action]. Rationale:
...

Recommendations may cover: user validation methods, MVP boundary adjustments, competitive strategy, metric design, team allocation, next milestone priorities.