jpskill.com
💼 ビジネス コミュニティ

knearme-sprint-workflow

Manage KnearMe portfolio development sprints. This skill should be used when starting a work session, deciding what to work on next, or tracking progress on sprint tasks. It provides guidance on sub-agent delegation and sprint file updates.

⚡ おすすめ: コマンド1行でインストール(60秒)

下記のコマンドをコピーしてターミナル(Mac/Linux)または PowerShell(Windows)に貼り付けてください。 ダウンロード → 解凍 → 配置まで全自動。

🍎 Mac / 🐧 Linux
mkdir -p ~/.claude/skills && cd ~/.claude/skills && curl -L -o knearme-sprint-workflow.zip https://jpskill.com/download/8665.zip && unzip -o knearme-sprint-workflow.zip && rm knearme-sprint-workflow.zip
🪟 Windows (PowerShell)
$d = "$env:USERPROFILE\.claude\skills"; ni -Force -ItemType Directory $d | Out-Null; iwr https://jpskill.com/download/8665.zip -OutFile "$d\knearme-sprint-workflow.zip"; Expand-Archive "$d\knearme-sprint-workflow.zip" -DestinationPath $d -Force; ri "$d\knearme-sprint-workflow.zip"

完了後、Claude Code を再起動 → 普通に「動画プロンプト作って」のように話しかけるだけで自動発動します。

💾 手動でダウンロードしたい(コマンドが難しい人向け)
  1. 1. 下の青いボタンを押して knearme-sprint-workflow.zip をダウンロード
  2. 2. ZIPファイルをダブルクリックで解凍 → knearme-sprint-workflow フォルダができる
  3. 3. そのフォルダを C:\Users\あなたの名前\.claude\skills\(Win)または ~/.claude/skills/(Mac)へ移動
  4. 4. Claude Code を再起動

⚠️ ダウンロード・利用は自己責任でお願いします。当サイトは内容・動作・安全性について責任を負いません。

🎯 このSkillでできること

下記の説明文を読むと、このSkillがあなたに何をしてくれるかが分かります。Claudeにこの分野の依頼をすると、自動で発動します。

📦 インストール方法 (3ステップ)

  1. 1. 上の「ダウンロード」ボタンを押して .skill ファイルを取得
  2. 2. ファイル名の拡張子を .skill から .zip に変えて展開(macは自動展開可)
  3. 3. 展開してできたフォルダを、ホームフォルダの .claude/skills/ に置く
    • · macOS / Linux: ~/.claude/skills/
    • · Windows: %USERPROFILE%\.claude\skills\

Claude Code を再起動すれば完了。「このSkillを使って…」と話しかけなくても、関連する依頼で自動的に呼び出されます。

詳しい使い方ガイドを見る →
最終更新
2026-05-18
取得日時
2026-05-18
同梱ファイル
1
📖 Claude が読む原文 SKILL.md(中身を展開)

この本文は AI(Claude)が読むための原文(英語または中国語)です。日本語訳は順次追加中。

KnearMe Sprint Workflow

Overview

This skill guides development of the KnearMe Portfolio platform through 6 sprints. It ensures sprint files stay updated, progress is tracked, and sub-agents are used effectively.

Sprint Files Location: knearme-portfolio/todo/

Starting a Work Session

  1. Check Current Sprint Status

    cat knearme-portfolio/todo/README.md | grep -A 12 "Sprint Overview"
  2. Read Active Sprint File

    • Current: knearme-portfolio/todo/sprint-1-foundation.md
    • Find incomplete tasks: grep -n "\[ \]" knearme-portfolio/todo/sprint-1-foundation.md | head -10
  3. Use TodoWrite Tool to track session tasks from sprint file

What to Work on Next

Priority Order

  1. Blockers - Tasks preventing other work
  2. Dependencies - Earlier numbered sections before later ones
  3. Critical path - Auth → AI → UX → SEO → Polish → Launch
  4. Quick wins - Small tasks that unblock multiple others

Finding Next Task

  1. Open current sprint file
  2. Find first unchecked [ ] in lowest incomplete section
  3. Verify prerequisites (earlier sections done)
  4. If blocked, note in "Notes" section and skip to next independent task

Updating Sprint Progress

During Work

  • Mark tasks complete immediately: [ ][x]
  • Add brief notes for decisions made
  • Note blockers preventing completion

End of Session

  1. Commit sprint file changes
  2. Update README.md if sprint status changes
  3. Note next recommended task for continuity

Sprint Completion

When all section checkboxes are done:

  1. Verify "Definition of Done" criteria met
  2. Update README.md: change status from 🔄 to ✅
  3. Move to next sprint

Sub-Agent Patterns

When to Use Each Agent Type

Agent Type Use For
Explore Understanding codebase, finding patterns, scoping work
Plan Architecture decisions, multi-file changes, complex features
feature-dev:code-architect Designing new features before implementation
feature-dev:code-reviewer After significant code changes
general-purpose Complex multi-step tasks

Sprint Task → Agent Mapping

Sprint Task Type Recommended Pattern
Database schema creation Use Supabase MCP apply_migration
New UI component feature-dev:code-architect → implement
New API route Explore existing patterns → implement → code-reviewer
Multi-file refactor Plan first → parallel implementation
Bug fix Explore root cause → fix → test

Parallel Agent Work

For independent sprint tasks, launch agents in parallel:

Agent 1: Backend/database tasks
Agent 2: Frontend components
Agent 3: Testing/documentation

Example: In Sprint 1, these can run in parallel:

  • Database schema setup (Supabase MCP)
  • Auth UI components (feature-dev)
  • Profile setup form (feature-dev)

Quick Reference Commands

# Check overall progress
grep -c "\[x\]" knearme-portfolio/todo/*.md   # Completed tasks
grep -c "\[ \]" knearme-portfolio/todo/*.md   # Remaining tasks

# Find next task in current sprint
grep -n "\[ \]" knearme-portfolio/todo/sprint-1-foundation.md | head -5

# Check current sprint status
head -20 knearme-portfolio/todo/README.md

# Run progress script
./.claude/skills/knearme-sprint-workflow/scripts/check_progress.sh

Sprint Dependency Chain

Sprint 1: Foundation & Auth    ← Start here
    ↓
Sprint 2: AI Pipeline          ← Needs auth working
    ↓
Sprint 3: Core UX              ← Needs AI endpoints
    ↓
Sprint 4: Portfolio & SEO      ← Needs content to display
    ↓
Sprint 5: Polish & PWA         ← Needs features to polish
    ↓
Sprint 6: Launch               ← Final deployment

Important Files

  • Sprint tracking: knearme-portfolio/todo/sprint-*.md
  • Technical docs: knearme-portfolio/CLAUDE.md
  • Contributor guide: knearme-portfolio/AGENTS.md
  • Product specs: knearme-portfolio/docs/02-requirements/epics/
  • Architecture: knearme-portfolio/docs/03-architecture/

Sprint Overview

Sprint Focus Key Deliverables
Sprint 1 Foundation & Auth Next.js setup, Supabase, Auth flows, Profile setup
Sprint 2 AI Pipeline GPT-4V analysis, Whisper transcription, GPT-4o generation
Sprint 3 Core UX Photo upload, Interview flow, Editing, Publishing
Sprint 4 Portfolio & SEO Public pages, Schema.org, Sitemap, Performance
Sprint 5 Polish & PWA Service worker, Accessibility, Testing
Sprint 6 Launch Production deploy, Monitoring, Onboarding

Session Handoff Pattern

At the end of each session, leave a clear handoff:

## Session Notes (YYYY-MM-DD)

### Completed
- [x] Task 1
- [x] Task 2

### In Progress
- Task 3 (started, needs X to finish)

### Blockers
- Need clarification on Y

### Next Up
- First incomplete task in sprint file

Add this to the sprint file's "Notes" section for continuity.