one-pager-prd
About
This Claude Skill generates concise product requirement documents (PRDs) for proposing features and aligning stakeholders. It creates decision-ready specs covering problems, solutions, users, and success metrics. Use it when scoping projects, pitching initiatives, or whenever product documentation like one-pagers or feature proposals are needed.
Documentation
One-Pager PRD
Table of Contents
Purpose
Create concise, decision-ready product specifications that align stakeholders on problem, solution, users, success metrics, and constraints—enabling fast approval and reducing back-and-forth.
When to Use
Early-Stage Product Definition:
- Proposing new feature or product
- Need stakeholder alignment before building
- Scoping initiative for resource allocation
- Pitching idea to leadership for approval
Documentation Needs:
- Capturing requirements for engineering handoff
- Documenting decisions for future reference
- Creating spec for cross-functional team (PM, design, eng)
- Recording what's in/out of scope
Communication:
- Getting buy-in from multiple stakeholders
- Explaining complex feature simply
- Aligning sales/marketing on upcoming releases
- Onboarding new team members to initiative
When NOT to Use:
- Detailed technical design docs (use ADRs instead)
- Comprehensive product strategy (too high-level for one-pager)
- User research synthesis (different format)
- Post-launch retrospectives (use postmortem skill)
What Is It
A one-pager PRD is a 1-2 page product specification covering:
Core Elements:
- Problem: What user pain are we solving? Why now?
- Solution: What are we building? (High-level approach)
- Users: Who benefits? Personas, segments, use cases
- Goals & Metrics: How do we measure success?
- Scope: What's in/out? Key user flows
- Constraints: Technical, business, timeline limits
- Open Questions: Unknowns to resolve
Format:
- One-Pager: 1 page, bullet points, for quick approval
- PRD (Product Requirements Document): 1-2 pages, more detail, for execution
Example One-Pager:
Feature: Bulk Edit for Data Tables
Problem: Users managing 1000+ rows waste hours editing one-by-one. Competitors have bulk edit. Churn risk for power users.
Solution: Select multiple rows → Edit panel → Apply changes to all. Support: text, dropdowns, dates, numbers.
Users: Data analysts (15% of users, 60% of usage), operations teams.
Goals: Reduce time-to-edit by 80%. Increase retention of power users by 10%. Launch Q2.
In Scope: Select all, filter+select, edit common fields (5 field types). Out of Scope: Undo/redo (v2), bulk delete (security concern).
Metrics: Time per edit (baseline: 5 min/row), adoption rate (target: 40% of power users in month 1).
Constraints: Must work with 10K rows without performance degradation.
Open Questions: Validation—fail entire batch or skip invalid rows?
Workflow
Copy this checklist and track your progress:
One-Pager PRD Progress:
- [ ] Step 1: Gather context
- [ ] Step 2: Choose format
- [ ] Step 3: Draft one-pager
- [ ] Step 4: Validate quality
- [ ] Step 5: Review and iterate
Step 1: Gather context
Identify the problem (user pain, data supporting it), proposed solution (high-level approach), target users (personas, segments), success criteria (goals, metrics), and constraints (technical, business, timeline). See Common Patterns for typical problem types.
Step 2: Choose format
For simple features needing quick approval → Use resources/template.md one-pager format (1 page, bullets). For complex features/products requiring detailed requirements → Use resources/template.md full PRD format (1-2 pages). For writing guidance and structure → Study resources/methodology.md for problem framing, metric definition, scope techniques.
Step 3: Draft one-pager
Create one-pager-prd.md with: problem statement (user pain + why now), solution overview (what we're building), user personas and use cases, goals with quantified metrics, in-scope flows and out-of-scope items, constraints and assumptions, open questions to resolve. Keep concise—1 page for one-pager, 1-2 for PRD.
Step 4: Validate quality
Self-assess using resources/evaluators/rubric_one_pager_prd.json. Check: problem is specific and user-focused, solution is clear without being overly detailed, metrics are measurable and have targets, scope is realistic and boundaries clear, constraints acknowledged, open questions identified. Minimum standard: Average score ≥ 3.5.
Step 5: Review and iterate
Share with stakeholders (PM, design, engineering, business). Gather feedback on problem framing, solution approach, scope boundaries, and success metrics. Iterate based on input. Get explicit sign-off before moving to detailed design/development.
Common Patterns
By Problem Type
User Pain (Most Common):
- Pattern: Users can't do X, causing friction Y
- Example: "Users can't search by multiple filters, forcing 10+ clicks to find items"
- Validation: User interviews, support tickets, analytics showing workarounds
Competitive Gap:
- Pattern: Competitor has feature, we don't, causing churn
- Example: "Competitors offer bulk actions. 20% of churned users cited this"
- Validation: Churn analysis, competitive analysis, win/loss interviews
Strategic Opportunity:
- Pattern: Market shift creates opening for new capability
- Example: "Remote work surge → need async collaboration features"
- Validation: Market research, early customer interest, trend analysis
Technical Debt/Scalability:
- Pattern: Current system doesn't scale, blocking growth
- Example: "Database queries timeout above 100K users. Growth blocked."
- Validation: Performance metrics, system capacity analysis
By Solution Complexity
Simple Feature (Weeks):
- Clear requirements, minor scope
- Example: Add export to CSV button
- One-Pager: Problem, solution (1 sentence), metrics, constraints
Medium Feature (Months):
- Multiple user flows, some complexity
- Example: Commenting system with notifications
- PRD: Detailed flows, edge cases, phasing (MVP vs v2)
Large Initiative (Quarters):
- Cross-functional, strategic
- Example: Mobile app launch
- Multi-Page PRD: Break into phases, each with own one-pager
By User Segment
B2B SaaS:
- Emphasize: ROI, admin controls, security, integrations
- Metrics: Adoption rate, time-to-value, NPS
B2C Consumer:
- Emphasize: Delight, ease of use, viral potential
- Metrics: Daily active users, retention curves, referrals
Enterprise:
- Emphasize: Compliance, customization, support
- Metrics: Deal size impact, deployment success, enterprise NPS
Internal Tools:
- Emphasize: Efficiency gains, adoption by teams
- Metrics: Time saved, task completion rate, employee satisfaction
Guardrails
Problem:
- Specific, not vague: ❌ "Users want better search" → ✓ "Users abandon search after 3 failed queries (30% of sessions)"
- User-focused: Focus on user pain, not internal goals ("We want to increase engagement" is goal, not problem)
- Validated: Cite data (analytics, interviews, support tickets) not assumptions
Solution:
- High-level, not over-specified: Describe what, not how. Leave design/engineering latitude.
- Falsifiable: Clear enough that stakeholders can disagree or suggest alternatives
- Scope-appropriate: Don't design UI in one-pager. "Filter panel" not "Dropdown menu with checkbox multi-select"
Metrics:
- Measurable: Must be quantifiable. ❌ "Improve UX" → ✓ "Reduce time-to-complete from 5 min to 2 min"
- Leading + Lagging: Include both (leading: adoption rate, lagging: revenue impact)
- Baselines + Targets: Current state + goal. "Increase NPS from 40 to 55"
Scope:
- Crisp boundaries: Explicitly state what's in/out
- MVP vs Future: Separate must-haves from nice-to-haves
- User flows: Describe key happy path, edge cases
Constraints:
- Realistic: Acknowledge tech debt, dependencies, timeline limits
- Trade-offs: Explicit about what we're sacrificing (speed vs quality, features vs simplicity)
Red Flags:
- Solution looking for problem (built it because cool tech, not user need)
- Vague metrics (no baselines, no targets, no timeframes)
- Scope creep (everything is "must-have")
- No constraints mentioned (unrealistic optimism)
- No open questions (haven't thought deeply)
Quick Reference
Resources:
resources/template.md- One-pager and PRD templates with section guidanceresources/methodology.md- Problem framing techniques, metric trees, scope prioritization, writing clarityresources/evaluators/rubric_one_pager_prd.json- Quality criteria
Output: one-pager-prd.md with problem, solution, users, goals/metrics, scope, constraints, open questions
Success Criteria:
- Problem is specific with validation (data/research)
- Solution is clear high-level approach (what, not how)
- Metrics are measurable with baselines + targets
- Scope has crisp in/out boundaries
- Constraints acknowledged
- Open questions identified
- Stakeholder sign-off obtained
- Score ≥ 3.5 on rubric
Quick Decisions:
- Simple feature, quick approval? → One-pager (1 page, bullets)
- Complex feature, detailed handoff? → Full PRD (1-2 pages)
- Multiple phases? → Separate one-pager per phase
- Strategic initiative? → Start with one-pager, expand to multi-page if needed
Common Mistakes:
- Problem too vague ("improve experience")
- Solution too detailed (specifying UI components)
- No metrics or unmeasurable metrics
- Scope creep (no "out of scope" section)
- Ignoring constraints (unrealistic timelines)
- No validation (assumptions not data)
- Writing for yourself, not stakeholders
Key Insight: Brevity forces clarity. If you can't explain it in 1-2 pages, you haven't thought it through. One-pager is thinking tool as much as communication tool.
Format Tips:
- Use bullets, not paragraphs (scannable)
- Lead with problem (earn the right to propose solution)
- Quantify everything possible (numbers > adjectives)
- Make scope boundaries explicit (prevent misunderstandings)
- Surface open questions (show you've thought deeply)
Stakeholder Adaptation:
- For Execs: Emphasize business impact, metrics, resource needs
- For Engineering: Technical constraints, dependencies, phasing
- For Design: User flows, personas, success criteria
- For Sales/Marketing: Competitive positioning, customer value, launch timing
Quick Install
/plugin add https://github.com/lyndonkl/claude/tree/main/one-pager-prdCopy and paste this command in Claude Code to install this skill
GitHub 仓库
Related Skills
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
llamaguard
OtherLlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
