Design Sprint Prototype Plan (Thursday morning)

Quick facts

Classification: tool | Version: 0.1.0 | Category: discovery | License: Apache-2.0

Try it: /tool-design-sprint-prototype-plan "Your context here"

Thursday is the prototype build day. This skill produces the morning planning artifact that makes the build run; it does NOT build the prototype itself. The skill assigns roles, locks the fidelity bar, drafts the Five-Act interview script, defines the trial-run gate, and confirms Friday participants. The actual building (Figma frames, slideware decks, role-play scripts, paper assemblies) is craft work the team owns through the rest of Thursday.

Family contract: docs/reference/skill-families/design-sprint-skills-contract.md. This skill is a member of design-sprint-skills.

When to Use

It is Thursday morning of the Design Sprint and Wednesday’s storyboard is signed off.
The Friday participant tracker needs a Thursday-morning confirmation check; cancellations now trigger buffer activation.
The interview script needs to be drafted (or refined from a prior sprint’s template) before the interviewer can mock-run it Thursday afternoon.
The team needs role assignments before parallel build work begins; ad-hoc assignment Thursday afternoon wastes 1-2 hours.

When NOT to Use

Wednesday is not closed. Return to tool-design-sprint-decide-and-storyboard; without a locked storyboard, the prototype has no spec.
The team is trying to use this skill to build the prototype. The prototype is craft work (per Ratified Decision 1 in the DS integration plan). This skill plans Thursday; the build happens in Figma, Keynote, paper, or whatever the medium is.
The team has decided to skip Friday’s trial run. Trial run is the gate that catches “prototype is unreachable” or “interview script confuses customer” before live interviews start. Skipping it converts Friday into a debugging session.
Friday participant cancellations have triggered cohort-shrinks-to-fewer-than-4. Pause the sprint and decide whether to postpone Friday or accept a reduced-evidence test.

How to Use

Use the /tool-design-sprint-prototype-plan slash command:

/tool-design-sprint-prototype-plan "Your context here"

Or reference the skill file directly: skills/tool-design-sprint-prototype-plan/SKILL.md

Output Template

Design Sprint Thursday Planning Artifact: [Initiative or Challenge name]

Prototype Role Plan

Role	Owner	Day allocation	Notes
Maker	[Name]	[Hours; typically 09:30-16:00]	[What surfaces they own]
Stitcher	[Name; may equal Maker on small teams]	[Hours; typically 13:00-16:30]	[Integration work]
Writer	[Name]	[Hours; typically 09:30-12:30]	[Copy ownership; reviewed by Interviewer for script consistency]
Asset Collector	[Name; may double up]	[Hours; flexible]	[Asset sources; rights/licensing posture]
Interviewer	[Name]	[Hours; 09:30-12:30 script draft + 14:00-15:00 mock-run + 15:30-17:00 trial run]	[Script ownership and trial-run lead]

Doubling-up notes: [If team is 4-person, which roles merge.]

Prototype Brief

What to build: [Storyboard panels in scope; reference the panel numbers from the Wednesday storyboard.]

Fidelity bar: [What “real enough” means for the chosen medium. One paragraph. Examples: for clickable Figma, “every storyboard panel has a frame and inter-frame interactions; copy is final-ish; visual design is on-brand but not pixel-perfect”; for slideware, “every storyboard panel is a slide; transitions are linear; participant cannot navigate freely.”]

Time allocation per role: [See role plan above; this section captures any constraint not in the plan, e.g., “Writer finishes copy by 12:30 PT so Stitcher has final copy for inter-frame links.”]

Explicitly NOT being built: [3-5 bullets. What the team agreed in Wednesday’s storyboard to defer (visual polish, animation timing, additional flow branches, additional surface details).]

Interview Script (Five-Act)

Act 1: Welcome (5 min)

[Canonical wording with team-name swap. Typical script: “Hi [name], thanks for joining today. I’m [interviewer], and I work on Brainshelf. The next 60 minutes will be a conversation, not a test - there are no wrong answers, and the prototype you’ll see is early, so some things won’t work the way you’d expect. I’ll ask you about your current habits, then show you something we’re working on, then ask what you think. Sound good?”]

Act 2: Context (5-10 min)

[Canonical questions adapted to the challenge. Typical questions:]

[Tell me about your typical week with [challenge domain, e.g., books]; what’s that like for you?]
[What do you do when [the challenge moment happens, e.g., you encounter a book you want to remember]?]
[What’s worked well or frustrated you about your current approach?]
[If you could change one thing about [the challenge moment], what would it be?]

Act 3: Intro (5 min)

[Canonical wording. Typical script: “Okay, now I’m going to show you something we’re working on. It’s an early prototype, so some things won’t work, and some screens may look unfinished. I’d love it if you’d think out loud as you go through it - just describe what you’re seeing, what you’d expect, and what you’re trying to do. Don’t worry about being polite about the design; honest reactions are exactly what we need.”]

Act 4: Tasks (20-30 min)

[Team-supplied wording derived from storyboard. Open-ended; not over-scripted. Typical 3 tasks for a 30-min Tasks act:]

[Task 1: open-ended primary task; e.g., “You just picked up [book name] at a bookstore. Try to add it to Brainshelf and find it later in your library.”]
[Task 2: secondary task that tests a non-primary storyboard panel; e.g., “It’s three weeks later. You’re thinking about a book you read recently about [topic]. Try to find it.”]
[Task 3: optional probe task; e.g., “Imagine you’ve used this for 6 months and have 50 books in your library. Show me what you’d do to find a book you barely remember.”]

Interviewer probing notes: [Wording the interviewer can use to probe without leading. Examples: “Walk me through what you’re thinking now.” / “What did you expect to happen?” / “If this wasn’t here, what would you try instead?”]

Act 5: Debrief (5-10 min)

[Canonical structure adapted for pricing/recall/etc. Typical questions:]

[What was your overall reaction to the prototype?]
[What worked well? What didn’t?]
[If you could change one thing about [the prototype’s primary moment], what would it be?]
[If this product were real, what would you expect to pay for it, and how often?]
[Anything else you want to share?]

[Close: “Thank you so much for your time. The honorarium will arrive via [method] within [days]. Have a great rest of your day.”]

Trial-Run Checklist

[Run with a fake customer (a teammate playing a target-profile customer) Thursday 16:00-17:00. Each gate is yes/no.]

Prototype reaches every storyboard panel without dead links or missing assets
All inter-panel interactions trigger as expected (Figma click targets, Keynote transitions, role-play cues)
Final copy is in place; no lorem ipsum, no “TODO”, no placeholder text
Asset Collector confirms all sample data (book titles, names, dates) is fake-but-believable; no real-customer data leaked
Interviewer can complete the full Five-Act script in 55-65 minutes (target 60 minutes including buffer)
Tasks act produces observable, probe-able behavior from the fake customer (not just script-following)
Recording setup (Zoom recording on; Otter or equivalent transcription armed; consent form sent)
Observer-room setup confirmed (breakout room or separate Zoom; sticky-note tool open)
Participant calendar invites confirmed for all 5 + 1 buffer slots; Zoom links sent

Participant Confirmation Tracker (Thursday Morning)

Slot	Customer (anonymized)	Status as of Thursday 09:30	Recruiter action
Fri 09:00	[Customer ID]	[Confirmed / Reminder sent / At risk]	[None / Send reminder / Activate buffer]
Fri 10:30	[Customer ID]	[Status]	[Action]
Fri 12:00	[Customer ID]	[Status]	[Action]
Fri 14:00	[Customer ID]	[Status]	[Action]
Fri 15:30	[Customer ID]	[Status]	[Action]
Fri 17:00 (buffer)	[Customer ID or “open”]	[Status]	[Action]

Risk assessment for Friday: [Low / Medium / High; cohort confidence; cancellation cascade risk]

Decider Checkpoint

Decider sign-off required before parallel build work begins (canonically by 10:00 PT Thursday).

Decider approves the role plan and confirms no role is unassigned.
Decider approves the fidelity bar; team will NOT polish past this bar.
Decider approves the Five-Act interview script’s Tasks act wording (the team-supplied portion).
Decider approves the trial-run gate criteria; trial run MUST pass before Friday begins.
Decider acknowledges participant tracker status; if at-risk slots exist, recruiter actions are authorized.
Decider confirms attendance for Friday PM 14:00-18:00 PT Decider window (observe interview slots 4-5 + Decider review concludes by 17:30 PT).

Signed: [Decider name, role], [ISO date and local time]

Example Output

Design Sprint Thursday Planning Artifact: Brainshelf Camera-Capture Validation

Thursday morning 2026-06-04, 09:30 PT. Wednesday closed with Sketch B supervoted all-in-one and a 9-panel storyboard locked. Today: parallel Figma build (Alex + Sam paired); Riley drafts and mock-runs the Five-Act interview script; Jamie reviews build progress and signs off on the plan before the day’s parallel work begins.

Prototype Role Plan

Role	Owner	Day allocation	Notes
Maker	Alex	09:30-16:00 PT	Owns all 9 Figma frames + visual design + interaction surfaces
Stitcher	Alex (paired with Sam)	13:00-16:30 PT	Connects Figma frames; Sam owns interaction-logic edge cases (5-sec undo timing, geolocation permission flow)
Writer	Jamie	09:30-12:30 PT	All prototype copy: button labels, system feedback strings, recall-surface search results copy; Riley reviews for interview-script consistency
Asset Collector	Sam (mornings) + Riley (afternoons)	09:30-15:00 PT	Source 6 fake-but-believable book covers (real books, real authors, but using public-domain or fair-use cover thumbnails); sample location data for “where you saw it” tags
Interviewer	Riley	09:30-12:30 script draft + 14:00-15:00 mock-run + 15:30-17:00 trial run	Solo on script Thursday morning; pulls in Sam as trial-run “customer” Thursday afternoon

Doubling-up notes: 4-person team. Alex covers Maker + Stitcher (single design owner; standard for first DS). Jamie covers Writer (PM has best context for product copy + brand voice). Sam + Riley share Asset Collector. Riley is dedicated Interviewer.

Prototype Brief

What to build: All 9 storyboard panels from Wednesday (Panel 1 Brainshelf icon tap through Panel 9 recall moment). Panels 1-5 are the load-bearing test panels (capture + confirmation + library view); Panels 6-9 are the contextual surround that lets the customer understand the recall payoff. Build all 9 as Figma frames; ensure Panels 1-5 have full interaction; Panels 6-9 may be lightly interactive (next-tap-advance only).

Fidelity bar: Will a customer believe this is real for the 30-minute interview? For Panels 1-5: every frame has a frame, every inter-frame interaction triggers within 200ms, copy is final, visual design is on-brand (Brainshelf navy + warm-white + serif headline; Inter for body), book cover thumbnails use real public-domain covers (no stock placeholder). For Panels 6-9: every frame exists; interactions are linear advance; copy is lightly polished but final (no TODOs or placeholders; Riley confirmed during trial run that nothing requires steering-around).

Time allocation per role: Per role plan above. Critical path: Writer finishes copy by 12:30 PT so Stitcher has final copy for Panel 3 and Panel 5 (the most copy-sensitive panels).

Explicitly NOT being built:

Animated transitions beyond the basic Figma smart-animate defaults (no custom timing curves)
Onboarding flow before Panel 1 (interviewer manually opens app to Panel 1 state)
Settings, profile, account screens (none of the storyboard tested moments require them)
The “books like this you have” recall surface from Sketch D (deferred to v0.2 per Wednesday Decider rationale)
Real OCR or geolocation (Figma simulates both with prepared book + prepared location)

Interview Script (Five-Act)

Act 1: Welcome (5 min)

Hi [name], thanks for joining today. I’m Riley, and I work on Brainshelf, an early-stage product for readers like you. The next 60 minutes will be a conversation, not a test. There are no wrong answers, and the prototype you’ll see is early, so some things won’t work the way you’d expect. I’ll ask you about your current habits with books, then show you something we’re working on, then ask what you think. I’ll be recording today’s session for our team to review later; that’s what the consent form covered. Sound good?

Act 2: Context (8 min)

Tell me about your typical reading life right now; how many books are you in or around, what kinds of books are catching your attention?
When you encounter a book you want to remember (at a bookstore, friend’s recommendation, online), what do you do?
What’s worked well or frustrated you about how you track or remember books today?
Think about a specific book you read in the last 2-3 years that you couldn’t remember the details of when you wanted to recall it. What happened?
If you could change one thing about how you remember books, what would it be?

Act 3: Intro (4 min)

Okay, now I’m going to show you something we’re working on. It’s an early prototype, so some things won’t work, and a couple of screens may look unfinished. I’d love it if you’d think out loud as you go through it. Just describe what you’re seeing, what you’d expect, what you’re trying to do. Don’t worry about being polite about the design; honest reactions are exactly what we need. The first thing you’ll see is the app open on a phone, with a book in your other hand. We’ll go from there.

Act 4: Tasks (28 min)

Task 1 (15 min): You just picked up a book called “The Overstory” at a bookstore called Twig and Cover Books. You want to remember you found it. Try to add it to Brainshelf.

Probing notes for Riley: “Walk me through what you’re thinking now.” “What did you expect to happen when you tapped that?” “If this didn’t work, what would you try instead?” Note timestamp when customer reaches Panel 4 (capture complete) to measure capture time.

Task 2 (8 min): It’s a week later. You want to find “The Overstory” in your library. Show me what you’d do.

Probing notes: “Where did you expect to see it?” “What does the ‘where you saw it’ tag tell you?” If customer never notices the geolocation tag, ask: “Did you see anything interesting on that card?” (after they’ve moved on).

Task 3 (5 min): Imagine you’ve used this for 6 months and have 50 books in your library. You’re thinking about something you read recently about old-growth forests. Show me what you’d do to find it.

Probing notes: This tests the recall payoff (Panel 8). Customers should be able to search “trees” or “forests” and find “The Overstory” via the captured-time context. If they don’t try search, prompt: “If you wanted to find it, where would you start?”

Act 5: Debrief (10 min)

What was your overall reaction to the prototype?
What worked well? What didn’t?
Did the capture flow feel fast enough to use without friction? Slow enough you’d give up on it?
The “where you saw it” tag - what did you make of that?
If this product were real, what would you expect to pay for it, and how often? (Probe: monthly subscription, one-time, ad-supported.)
Anything else you want to share?

Close: Thank you so much for your time. The USD 100 honorarium will arrive via [UserInterviews panel customers: UserInterviews credit; Discord customers: Amazon gift card] within 5 business days. Have a great rest of your day.

Trial-Run Checklist

Run with Sam playing a target-profile customer at 15:30-17:00 PT.

Participant Confirmation Tracker (Thursday Morning 09:30 PT)

Slot	Customer (anonymized)	Status as of Thursday 09:30	Recruiter action
Fri 09:00	UI-Panel-A (US, age 38, M)	Confirmed; reminder sent yesterday	None
Fri 10:30	Discord-1 (UK, age 31, F)	Confirmed; high enthusiasm	None
Fri 12:00	Discord-4 (US, age 36, M; moved up from buffer)	Confirmed; reminder sent	None
Fri 14:00	Discord-3 (Canada, age 29, M)	Confirmed	None
Fri 15:30	UI-Panel-B (Australia, age 52, F)	Confirmed; timezone 09:30 their local; comfortable confirmed	None
Fri 17:00 (buffer)	Open (previous buffer used Tuesday)	None scheduled	Riley to identify backup candidate by 12:00 PT Thursday in case of further cancellation

Risk assessment for Friday: Low. All 5 primary slots confirmed Thursday morning. Buffer slot empty but Riley has a backup-of-backup candidate identified (UI-Panel-C, age 41, F, US) who can be activated within 4 hours if needed.

Decider Checkpoint

Decider sign-off required before parallel build work begins (target 10:00 PT Thursday).

Jamie approves the role plan; Alex covers Maker + Stitcher; Sam + Riley share Asset Collector; Jamie covers Writer; Riley dedicated Interviewer.
Jamie approves the fidelity bar: Panels 1-5 full-fidelity; Panels 6-9 frame-with-linear-advance; no polishing past the bar.
Jamie approves the Five-Act Tasks act wording; specifically the 3-task structure: capture-The-Overstory / find-it-1-week-later / recall-old-growth-forests-6-months-later.
Jamie approves the trial-run gate criteria; if any gate fails at 17:00 PT, the team has until 19:00 PT to recover before Friday begins, otherwise Friday postpones.
Jamie acknowledges participant tracker status: 5 of 5 confirmed; buffer slot is open; Riley has backup-of-backup identified.
Jamie confirms attendance for Friday 14:00-18:00 PT Decider window (Decider observes interview slots 09:00-16:30 from breakout room; Decider review begins after 15:30 interview wraps 16:30 and concludes by 17:30 PT).

Signed: Jamie (founder, PM), 2026-06-04 09:55 PT.

Build authorized. Alex begins Figma build at 10:00 PT; trial run scheduled 15:30 PT; full plan locked.