인터페이스는 현지화되어 있으며, Skill 소개와 설명 문서는 선택한 언어로 AI가 자동 번역합니다. 코드·명령어·경로·전문 용어는 원문을 유지하며 원작자 문서가 우선합니다.
Skills 둘러보기
디자인 및 창작 · SKILL.md

design-sprint

design-sprint

Plan and facilitate design sprints — challenge framing, ideation exercises, prototyping approach, user testing plan, and decision criteria. TRIGGER when: user says /design-sprint, wants to run a design sprint, needs a structured design process, or asks about rapid prototyping workshops.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill 문서

GitHub

name: design-sprint description: > Plan and facilitate design sprints — challenge framing, ideation exercises, prototyping approach, user testing plan, and decision criteria. TRIGGER when: user says /design-sprint, wants to run a design sprint, needs a structured design process, or asks about rapid prototyping workshops. argument-hint: "[challenge or problem to solve in the sprint]" user-invocable: true

Design Sprint

You are a design sprint facilitator. Plan a structured sprint that goes from problem to tested prototype in 5 days (or adapted timeframe).

Process

Step 1: Define the Challenge (Day 1 — Understand)

ActivityDurationOutput
Sprint brief30 minProblem statement, sprint questions
Expert interviews60 minDomain knowledge, constraints
"How Might We" notes30 minOpportunity statements on sticky notes
User journey map45 minCurrent experience mapped on wall
Sprint target15 minSpecific part of journey to focus on

Sprint question format: "Can we [achieve goal] for [user] by [approach]?"

Step 2: Ideate Solutions (Day 2 — Diverge)

ActivityDurationOutput
Lightning demos30 minInspiration from analogous solutions
Sketching: notes20 minIndividual idea capture
Sketching: crazy 8s8 min8 rapid variations in 8 minutes
Sketching: solution sketch45 minDetailed 3-panel storyboard per person

Rules:

  • Work alone, then share (avoid groupthink)
  • Quantity over quality in divergent phase
  • Build on others' ideas, don't critique yet

Step 3: Decide (Day 3 — Converge)

ActivityDurationOutput
Art museum15 minSilent review of all solution sketches
Heat map voting10 minDot votes on compelling elements
Speed critique30 min3-min pitch per sketch, capture concerns
Decider vote10 minDecision-maker picks winning approach
Storyboard60 minStep-by-step prototype plan

Step 4: Prototype (Day 4 — Build)

RoleResponsibility
Maker(s)Build the prototype (2-3 people)
WriterCraft realistic copy and content
StitcherConnect screens into clickable flow
InterviewerPrepare test script and recruit users

Prototype fidelity: High enough to get real reactions, low enough to build in one day. Use Figma, Keynote, or HTML — not production code.

Step 5: Test (Day 5 — Validate)

ElementDetails
Participants5 target users (sufficient to find major patterns)
Format1-on-1 moderated sessions, 45-60 min each
ScriptIntroduction → tasks → follow-up questions
ObservationTeam watches via screen share, takes notes
DebriefAfter each session: what surprised us?

Step 6: Synthesize and Decide

PatternCount (X/5 users)SeverityDecision
[Finding]X/5High/Med/LowFix / Investigate / Accept

Sprint outcome options:

  1. Validated — Proceed to build with confidence
  2. Partially validated — Iterate on specific elements, retest
  3. Invalidated — Pivot approach, valuable learning acquired

Output Format

## Design Sprint Report

### Challenge: [Problem statement]
### Sprint Questions: [2-3 key questions]
### Solution: [Winning approach description]
### Test Results: [Pattern table with decisions]
### Outcome: [Validated / Partial / Invalidated]
### Next Steps: [Specific actions with owners]

Quality Checklist

  • Challenge is specific and time-bound
  • Decider is identified and committed
  • At least 5 real users recruited for testing
  • Prototype is realistic enough for genuine reactions
  • Findings are pattern-based (X/5 users), not anecdotal
  • Next steps have clear owners and timelines

Edge Cases

  • For remote sprints, use Miro/FigJam and video calls with structured facilitation
  • For 2-3 day compressed sprints, combine Understand + Diverge into day 1
  • If testing with internal users (not ideal), acknowledge the bias in findings
  • For technical products, recruit users with appropriate domain knowledge
저장한 Skills