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.