user-story-splitting
关于
This skill systematically breaks down large user stories or epics into smaller, deliverable units using proven split patterns. It's used when backlog items are too large for estimation, sequencing, or independent release. This makes work more manageable, reduces risk, and enables faster feedback cycles in agile development.
快速安装
Claude Code
推荐npx skills add deanpeters/Product-Manager-Skills -a claude-code/plugin add https://github.com/deanpeters/Product-Manager-Skillsgit clone https://github.com/deanpeters/Product-Manager-Skills.git ~/.claude/skills/user-story-splitting在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
Purpose
Break down large user stories, epics, or features into smaller, independently deliverable stories using systematic splitting patterns. Use this to make work more manageable, reduce risk, enable faster feedback cycles, and maintain flow in agile development. This skill applies to user stories, epics, and any work that's too large to complete in a single sprint.
This is not arbitrary slicing—it's strategic decomposition that preserves user value while reducing complexity.
Key Concepts
The Story Splitting Framework
Based on Richard Lawrence and Peter Green's "Humanizing Work Guide to Splitting User Stories," the framework provides 8 systematic patterns for splitting work:
- Workflow steps: Split along sequential steps in a user's journey
- Business rule variations: Split by different rule scenarios (permissions, calculations, etc.)
- Data variations: Split by different data types or inputs
- Acceptance criteria complexity: Split when multiple "When" or "Then" statements exist
- Major effort: Split by technical milestones or implementation phases
- External dependencies: Split along dependency boundaries (APIs, third parties, etc.)
- DevOps steps: Split by deployment or infrastructure requirements
- Tiny Acts of Discovery (TADs): When none of the above apply, use small experiments to unpack unknowns
Why Split Stories?
- Faster feedback: Smaller stories ship sooner, allowing earlier validation
- Reduced risk: Less to build = less that can go wrong
- Better estimation: Small stories are easier to estimate accurately
- Maintain flow: Keeps work moving through the sprint without bottlenecks
- Testability: Smaller scope = easier to write and run tests
Anti-Patterns (What This Is NOT)
- Not horizontal slicing: Don't split into "front-end story" and "back-end story" (each story should deliver user value)
- Not task decomposition: Stories aren't tasks ("Set up database," "Write API")
- Not arbitrary chopping: Don't split "Add user management" into "Add user" and "Management" (meaningless)
When to Use This
- Story is too large for a single sprint
- Multiple "When" or "Then" statements in acceptance criteria
- Epic needs to be broken down into deliverable increments
- Team can't reach consensus on story size or scope
- Story has multiple personas or workflows bundled together
When NOT to Use This
- Story is already small and well-scoped (don't over-split)
- Splitting would create dependencies that slow delivery
- The story is a technical task (use engineering task breakdown instead)
Application
Step 1: Identify the Original Story
Start with the story/epic/feature that needs splitting. Ensure it's written using the user story format (reference skills/user-story/SKILL.md or skills/epic-hypothesis/SKILL.md).
### Original Story:
[Story formatted with use case and acceptance criteria]
Step 2: Apply the Splitting Logic
Use template.md for the full fill-in structure and output format.
Work through the 8 splitting patterns in order. Stop when you find one that applies.
Pattern 1: Workflow Steps
Ask: Does this story contain multiple sequential steps?
Example:
- Original: "As a user, I want to sign up, verify my email, and complete onboarding"
- Split:
- "As a user, I want to sign up with email/password"
- "As a user, I want to verify my email via a confirmation link"
- "As a user, I want to complete onboarding by answering 3 profile questions"
Pattern 2: Business Rule Variations
Ask: Does this story have different rules for different scenarios?
Example:
- Original: "As a user, I want to apply discounts (10% for members, 20% for VIPs, 5% for first-time buyers)"
- Split:
- "As a member, I want to apply a 10% discount at checkout"
- "As a VIP, I want to apply a 20% discount at checkout"
- "As a first-time buyer, I want to apply a 5% discount at checkout"
Pattern 3: Data Variations
Ask: Does this story handle different types of data or inputs?
Example:
- Original: "As a user, I want to upload files (images, PDFs, videos)"
- Split:
- "As a user, I want to upload image files (JPG, PNG)"
- "As a user, I want to upload PDF documents"
- "As a user, I want to upload video files (MP4, MOV)"
Pattern 4: Acceptance Criteria Complexity
Ask: Does this story have multiple "When" or "Then" statements?
Example:
- Original: "As a user, I want to manage my cart"
- When I add an item, Then it appears in my cart
- When I remove an item, Then it disappears from my cart
- When I update quantity, Then the cart total updates
- Split:
- "As a user, I want to add items to my cart so I can purchase them later"
- "As a user, I want to remove items from my cart so I can change my mind"
- "As a user, I want to update item quantities so I can buy the right amount"
Note: This is the most common indicator that a story needs splitting. If you see multiple "When/Then" pairs, split along those boundaries.
Pattern 5: Major Effort
Ask: Does this story require significant technical work that can be delivered incrementally?
Example:
- Original: "As a user, I want real-time collaboration on documents"
- Split:
- "As a user, I want to see who else is viewing the document (read-only presence)"
- "As a user, I want to see live cursor positions of other editors"
- "As a user, I want to see live edits from other users in real-time"
Pattern 6: External Dependencies
Ask: Does this story depend on multiple external systems or APIs?
Example:
- Original: "As a user, I want to log in with Google, Facebook, or Twitter"
- Split:
- "As a user, I want to log in with Google OAuth"
- "As a user, I want to log in with Facebook OAuth"
- "As a user, I want to log in with Twitter OAuth"
Pattern 7: DevOps Steps
Ask: Does this story require complex deployment, infrastructure, or operational work?
Example:
- Original: "As a user, I want to upload large files to cloud storage"
- Split:
- "As a user, I want to upload small files (<10MB) to cloud storage"
- "As a user, I want to upload large files (10MB-1GB) with progress tracking"
- "As a user, I want to resume interrupted uploads"
Pattern 8: Tiny Acts of Discovery (TADs)
Ask: If none of the above apply, are there unknowns or assumptions that need unpacking?
Example:
- Original: "As a user, I want AI-powered recommendations" (too vague, too many unknowns)
- TADs:
- Prototype 3 recommendation algorithms and test with 10 users
- Define success criteria (click-through rate, user satisfaction)
- Build the simplest recommendation engine (collaborative filtering)
- Measure and iterate
Note: TADs aren't stories—they're experiments. Use them to de-risk and clarify before writing stories.
Step 3: Write the Split Stories
For each split, write a complete user story using the format from skills/user-story/SKILL.md:
### Split 1 using [Pattern Name]:
#### User Story [ID]:
- **Summary:** [Brief title]
**Use Case:**
- **As a** [persona]
- **I want to** [action]
- **so that** [outcome]
**Acceptance Criteria:**
- **Scenario:** [Description]
- **Given:** [Preconditions]
- **When:** [Action]
- **Then:** [Outcome]
Step 4: Validate the Splits
Ask these questions:
- Does each split deliver user value? (Not just "front-end done")
- Can each split be developed independently? (No hard dependencies)
- Can each split be tested independently? (Clear acceptance criteria)
- Is each split small enough for a sprint? (1-5 days of work)
- Do the splits, when combined, equal the original? (Nothing lost in translation)
If any answer is "no," revise.
Examples
See examples/sample.md for full splitting examples.
Mini example excerpt:
### Original Story:
As a team admin, I want to manage team members so that I can control access.
### Suggested Splits (Acceptance Criteria Complexity):
1. Invite new team members
2. Remove team members
3. Update team member roles
Common Pitfalls
Pitfall 1: Horizontal Slicing (Technical Layers)
Symptom: "Story 1: Build the API. Story 2: Build the UI."
Consequence: Neither story delivers user value independently.
Fix: Split vertically—each story should include front-end + back-end work to deliver a complete user-facing capability.
Pitfall 2: Over-Splitting
Symptom: "Story 1: Add button. Story 2: Wire button to API. Story 3: Display result."
Consequence: Creates unnecessary overhead and dependencies.
Fix: Only split when the story is too large. A 2-day story doesn't need splitting.
Pitfall 3: Meaningless Splits
Symptom: "Story 1: First half of feature. Story 2: Second half of feature."
Consequence: Arbitrary splits that don't map to user value or workflow.
Fix: Use one of the 8 splitting patterns—each split should have a clear rationale.
Pitfall 4: Creating Hard Dependencies
Symptom: "Story 2 can't start until Story 1 is 100% done, tested, and deployed."
Consequence: No parallelization, slows delivery.
Fix: Split in a way that allows independent development. If dependencies are unavoidable, prioritize Story 1.
Pitfall 5: Ignoring the "So That"
Symptom: Split stories have the same "so that" statement.
Consequence: You've split the action but not the outcome—likely a task decomposition, not a story split.
Fix: Ensure each split has a distinct user outcome. If not, reconsider the split pattern.
References
Related Skills
skills/user-story/SKILL.md— Format for writing the split storiesskills/epic-hypothesis/SKILL.md— Epics often need splitting before becoming storiesskills/jobs-to-be-done/SKILL.md— Helps identify meaningful splits along user jobs
External Frameworks
- Richard Lawrence & Peter Green, The Humanizing Work Guide to Splitting User Stories — Origin of the 8 splitting patterns
- Bill Wake, INVEST in Good Stories (2003) — Criteria for well-formed stories (Independent, Negotiable, Valuable, Estimable, Small, Testable)
- Mike Cohn, User Stories Applied (2004) — Story decomposition techniques
Dean's Work
- User Story Splitting Prompt Template (based on Humanizing Work framework)
Provenance
- Adapted from
prompts/user-story-splitting-prompt-template.mdin thehttps://github.com/deanpeters/product-manager-promptsrepo.
Skill type: Component
Suggested filename: user-story-splitting.md
Suggested placement: /skills/components/
Dependencies: References skills/user-story/SKILL.md, skills/epic-hypothesis/SKILL.md
Applies to: User stories, epics, and any work that's too large to complete in a single sprint
GitHub 仓库
相关推荐技能
llamaguard
其他LlamaGuard是Meta推出的7-8B参数内容审核模型,专门用于过滤LLM的输入和输出内容。它能检测六大安全风险类别(暴力/仇恨、性内容、武器、违禁品、自残、犯罪计划),准确率达94-95%。开发者可通过HuggingFace、vLLM或Sagemaker快速部署,并能与NeMo Guardrails集成实现自动化安全防护。
cost-optimization
其他这个Claude Skill帮助开发者优化云成本,通过资源调整、标记策略和预留实例来降低AWS、Azure和GCP的开支。它适用于减少云支出、分析基础设施成本或实施成本治理策略的场景。关键功能包括提供成本可视化、资源规模调整指导和定价模型优化建议。
quantizing-models-bitsandbytes
其他这个Skill使用bitsandbytes库量化大语言模型,能在GPU内存有限时通过8位或4位量化减少50-75%内存占用,同时保持精度损失最小。它支持INT8、NF4、FP4等多种量化格式,可与HuggingFace Transformers无缝集成,适用于需要部署更大模型或加速推理的场景。还提供QLoRA训练和8位优化器支持,让开发者能轻松实现高效模型压缩。
dispatching-parallel-agents
其他该Skill用于并行处理3个以上无依赖关系的独立故障,可为每个问题域分派专属Claude代理同时执行调查修复。它通过并发处理多个独立问题显著提升故障排查效率,特别适用于测试文件、子系统等无共享状态的场景。
