無料GitHub で公開

epic-breakdown

Decompose large epics into sprint-ready sub-epics using 9 splitting patterns. Use this skill when: - A spec is approved and you need to break a large feature into estimable work items for engineering - You want to split an epic that is too large for a single sprint - You need sized sub-epics with story point estimates for sprint planning - You are preparing for engineering handoff and need to identify work packages

qa-aman/claude-skills

インストール方法を見る

含まれるファイル（2）

SKILL.md3.7 KB
references/splitting-patterns.md3.2 KB

SKILL.md（原文）

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Epic Breakdown

Take a spec's Functional Requirements and Differences sections, apply 9 decomposition patterns to produce sized sub-epics ready for sprint planning.

Workflow

Read spec from your spec directory
List all functionality blocks and platform/user differences
Evaluate which splitting patterns apply to each block
Generate sub-epics with estimates
Map dependencies between sub-epics
Suggest sprint allocation
Output to file or print

The 9 Splitting Patterns

#	Pattern	When to Use
1	Workflow Steps	Multi-step flow (create -> configure -> publish)
2	Business Rules	Complex validation logic
3	Data Variations	Multiple content types or input formats
4	Platform Differences	Different behavior across web/mobile/tablet/offline
5	Operations (CRUD)	Standard data management flow
6	Spike/Research	Technical unknowns or third-party integrations
7	Performance	Specific NFR targets
8	Persona Variations	Different behavior per user role
9	Happy Path vs Edge Cases	Feature can deliver value with just happy path

See references/splitting-patterns.md for detailed examples of each pattern.

Story Point Guide

Points	Complexity	Example
1	Simple, no unknowns	Config change, copy update
2	Small, minor complexity	Add a field, simple UI change
3	Medium, some complexity	New API endpoint, form with validation
5	Significant, multiple components	New screen with backend integration
8	Large, cross-cutting	Offline sync, real-time updates
13	Very large — SPLIT FURTHER	Use the 9 patterns to decompose

If estimate > 8 points, attempt to split further using the patterns.

Output Format

# Epic Breakdown — {Feature Name}

**Source Spec:** {path}
**Date:** {DD-MM-YYYY}
**Total Sub-Epics:** {count}
**Total Estimated Points:** {sum}

---

## Sub-Epics

### SE-001: {Title}
**Pattern:** {which of the 9}
**Estimate:** {points} points
**Spec Section:** {reference}
**Description:** {what this delivers}
**Dependencies:** {SE-NNN or "None"}
**Acceptance Summary:** {key outcomes}

---

## Dependency Map

```mermaid
graph TD
  SE-001 --> SE-003
  SE-002 --> SE-003

Sprint Allocation

Sprint	Sub-Epics	Points	Theme
Sprint 1	SE-001, SE-002	{n}	Core happy path
Sprint 2	SE-003, SE-004	{n}	Edge cases + offline

Summary

Pattern Used	Count	Total Points
Workflow Steps	{n}	{n}
Happy Path vs Edge	{n}	{n}


## Anti-Patterns

- Don't create sub-epics smaller than 1 point — group them
- Don't ignore dependencies — they determine sprint order
- Don't estimate > 13 points — split further instead
- Don't skip the dependency map — it prevents sprint planning surprises

## Quality Checklist

- [ ] Every sub-epic has a clear pattern attribution
- [ ] All estimates are Fibonacci (1, 2, 3, 5, 8, 13)
- [ ] Dependencies are mapped
- [ ] Sprint allocation respects dependencies
- [ ] No sub-epic exceeds 8 points without a split rationale

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

ログインしてレビューを書く

同じリポジトリのスキル

概要と使いどころ

11-star-framework

無料

Rate any product, feature, or experience on the 11-star scale (Brian Chesky's Airbnb thought experiment). Use when user says "rate this experience", "11-star", "star rating", "experience audit", "how good is this", "experience rating", "product audit", "quality assessment", or wants to evaluate product quality and identify improvement paths. Also trigger when user wants to benchmark a feature, assess where a product stands, or map out what "great" looks like - even if they don't explicitly say "11-star".

日本語の概要は準備中です。原文の説明を表示しています。