Sprint Deck Outline Builder

Sprint Deck Outline Builder

Purpose

The written marketing plan (see sprint-marketing-plan-builder) is a document the client keeps. This skill produces something different: a slide-by-slide outline for presenting the sprint back to the client — usually in a wrap-up or kickoff-of-execution meeting. It's a narrative, not a status report: it opens on the client's own pain (why they came to the agency in the first place), moves through the plan that resolves it organized by Square, and closes by explicitly tying back to that opening pain — "here's how this sprint solves it" — before pointing toward continued engagement. Sprint logistics (dates, format, who facilitated, who was in the room) are not part of this narrative — they're context for the CMO presenting it, not something the client needs on a slide.

This is not a facilitation skill and it doesn't run a session. It assembles a presentation outline from work that's already done. If a Square wasn't run or a decision wasn't made, this skill's job is to say so on the slide — not to quietly invent a professional-looking slide that papers over the gap.

Core Rule

Same "nothing invented" standard as sprint-marketing-plan-builder, applied per slide: every Core content bullet on every slide must trace to a specific input — a stored Square output, or something identifiable in the source notes/transcript. Where the source is silent, the slide says so in brackets rather than filling the space with generic content. A deck that looks complete but quietly invents a decision, a date, or a "typical Square 7" description is worse than a deck with a visible gap — it tells the client, in front of the room, that something was decided when it wasn't. This applies to the opening pain slide and closing resolution slide just as much as any Square slide — the pain points must be real things the client said or that are identifiable in the source material, not a generic "businesses like yours struggle with X" template.

One deliberate exception: the Speaker notes field on each slide may include facilitation and delivery guidance — how to present the slide, what tone to strike, where to pause for questions — without needing a transcript citation. Notes are about how to deliver the material, not what the material is. The distinction that matters: Core content is a claim about what THIS client decided, and needs a source; speaker notes are presentation guidance, and don't. Don't let that exception become a loophole for smuggling invented substance into the notes field instead of the content field — if a speaker note asserts a fact ("mention that the team agreed to launch by September"), that fact still needs to trace to source just like a bullet would. Sprint logistics (date, format, facilitators, who was in the room) belong here now, in the opening slide's speaker notes, rather than on their own slide — see Deck Structure below. Persona methodology notes (how a persona was scored or chosen, which alternative segments were considered and not adopted) also belong in speaker notes now rather than their own box on the persona slide — see the Square 1 guidance in Deck Structure. Square 2 and Square 3's own process/status detail follow the same pattern — see the Square 2/3 guidance below.

Inputs

Same inputs as sprint-marketing-plan-builder: stored Square skill outputs, raw sprint notes/transcripts, or both.

This skill is fully self-sufficient and does not depend on sprint-marketing-plan-builder existing or having been run — if that skill isn't available in this environment, or the deck is the only deliverable being requested, derive everything directly from the stored Square outputs or raw source using the reading strategy and rubric below. Every "same as the plan document's X section" instruction later in this file (objectives, implementation plan) describes what to derive and how, not a hard requirement that the other document be produced first.

The one case where reuse matters: if sprint-marketing-plan-builder has already been run against the same source material earlier in this session, reuse its Square-by-square extraction rather than re-deriving it from scratch. These two deliverables are commonly requested together — a client wrap-up often needs both the document and the deck from the same sprint. Re-parsing the same raw transcript a second time to re-derive what's already been extracted is wasted work and risks the two deliverables drifting out of sync with each other (e.g. one says a decision was "aligned," the other says it's still "open," because the second pass read the notes slightly differently). Pull from the plan builder's extraction directly when it's available in-session; otherwise derive from raw source using the process below — that's the normal path, not a fallback to apologize for.

If it's unclear which source material or which client's sprint this is, confirm before building — same as the plan builder.

Handling Raw Source Material

If working from raw notes/transcripts rather than a prior extraction, follow the same reading strategy as sprint-marketing-plan-builder: chapters typically break into Summary, a Decisions block (Aligned / Needs Further Discussion / Disagreements / Shelved — varies by sprint), Next steps, and a timestamped Details section, followed by a full transcript. Read Decisions → Details → Next steps first; use the transcript only to verify or resolve ambiguity. And map decisions to Squares by substance, not by looking for the word "Square" — it won't appear. Use the same rubric from the plan builder:

Preserve each decision's status exactly as sourced when it lands on a slide — an "aligned" bullet and an item still "needs further discussion" should never look the same on the page. (A still-open item that used to get its own Open Decisions slide now surfaces in speaker notes instead — see Deck Structure below.)

While reading, also pull out anything that describes the client's situation before this engagement — what was costing them money, what wasn't working, what prompted them to seek help, any frustration or risk they named. This is the raw material for the opening Pain slide (see Deck Structure) — it usually surfaces in the transcript's Summary, in Square 1/2 discussion of the customer's own pain (which sits alongside the client's), or in whatever framing the client gave for why they engaged the agency in the first place. Don't confuse the client's own pain (why this particular client sought out the agency) with their end customer's pain (e.g. Square 1/2 persona pain points) — they're related but distinct, and the opening slide is about the former. This same "client's own situation" material is also the source for the Implementation Plan slide's "What's at Stake" line — see below.

Keep It Simple — the BrandScript Principle

A slide is not a report. Think of each Square slide as a plain restatement of the one or two things that were actually decided — the kind of clean, obvious simplicity a BrandScript one-pager has — not a comprehensive recap of everything that was discussed, brainstormed, or mentioned in passing.

Concretely: 2-4 bullets per slide, each one short enough to read in a few seconds from the back of a room. If a Square produced a lot of material, that's a sign to pick the two or three decisions that actually matter and let the rest go, not a sign the slide needs more bullets. Cut brainstormed-but-not-decided ideas unless one is genuinely worth flagging as an open item — and if so, that's one bullet, not a paragraph. Square 2 and Square 3 push this principle furthest — see below, both reduce to a plain list as their primary content.

Per-Slide Template

Every slide in the outline follows this format, no exceptions:

Slide title

• Core content — 2-4 short bullets pulled directly from source, per the simplicity principle above. State the decision plainly; don't add interpretation, editorializing, or generic marketing commentary on top of it.
• Speaker notes — 1-3 sentences on what to actually say when presenting this slide, written so that someone who wasn't in the room for that part of the sprint can present it competently. Keep these tight too — a short cue, not a script.

Square 1 Gets One Slide Per Persona, Structured on the Buyer Persona Institute's Four Points

Square 1 is a deliberate exception to "one slide per Square" (step 4 in Deck Structure below): when the sprint identified more than one persona (a primary persona plus one or more secondary/deprioritized personas), each persona gets its own dedicated slide rather than combining them onto a single crowded slide. A primary-plus-secondary combo slide, or a slide that mixes persona description with scoring methodology and rejected alternatives, dilutes the thing the client actually needs to walk away remembering: who their customer is.

Each persona slide is structured around these four categories, adapted from the Buyer Persona Institute framework — use these exact four, in this order, not a different mix:

1. Pain points — what this persona struggles with that the client's product/service addresses.
2. Desired outcomes — what this persona is actually trying to achieve.
3. Barriers to investment — what holds this persona back from buying (price sensitivity, trust, timing, competing priorities, etc.).
4. Purchase criteria / questions — what this persona needs to know or see before they'll commit; the questions they're actually asking.

Source each of the four from the transcript/Square 1 output the same way as any other Core content — if the sprint's discussion of a given persona didn't cover one of the four categories in enough depth to state it confidently, that category gets its own yellow-flagged card on the slide (see the yellow flag rules below) rather than being invented or silently dropped.

What doesn't belong on the persona slide itself: methodology notes (how the persona was scored, e.g. "no formal PVP Index was run this session"), and mentions of alternative personas/segments that were considered and explicitly not adopted. These read as process detail, not something the client needs staring at them on a persona slide — move methodology notes into that slide's speaker notes. If a genuinely unresolved "should we reconsider this segment later" question exists, it also goes in that persona's speaker notes now (the deck no longer carries a standalone Open Decisions slide — see Deck Structure), or into the written marketing plan document's own Open Decisions section when one exists this session. If an alternative segment was cleanly rejected (not an open question), a single speaker-note line is enough; it doesn't need a slide, a box, or a mention anywhere else.

If only one persona came out of Square 1, this reduces to the normal single Square 1 slide, still structured on the same four categories rather than a generic "here's our customer" free-form layout.

Square 2 and Square 3 Reduce to a Plain List as Their Primary Content

Square 2 and Square 3 are the other two deliberate exceptions to a generic "recap everything this Square touched" slide — direct client feedback flagged both as reading better when they're simplified to the one list the client actually needs, with everything else moved out of Core content.

Square 2 (Marketing Message). Core content is a plain, flat list of the confirmed key messages/pillars — nothing else mixed in. If the sprint produced a Storybrand brand story for the client, that goes on this slide too (as a short narrative callout alongside or beneath the message list), since it's part of "what we say" just as much as the pillars are — source it from the transcript/Square 2 output the same as any Core content, and flag it yellow rather than fabricating one if the sprint didn't actually produce it. Everything else Square 2 typically touches — Brand Positioning Statement draft status, brand voice/tone alignment notes, naming or scope decisions (e.g. "keep the existing name rather than rebrand") — moves to this slide's speaker notes instead of sitting as Core bullets next to the message list. Those are process/status details, not the message itself, and diluting the list with them is exactly what the feedback was pushing back on.

Square 3 (Marketing Channels). Core content is a plain list of channels, split into two clearly labeled groups — Rented (paid/borrowed reach: ads, influencer placements, sponsored placements, etc.) and Owned (channels the client controls outright: email list, in-person/local events, their own content, etc.) — each channel with an optional one-line approach note if the source supports one (e.g. "YouTube — long-form education, not paid yet"). Sequencing or pacing decisions (e.g. "paid spend paused until the tech stack and messaging are locked") and anything that reads as an execution/commercial decision rather than a plain channel list move to the Implementation Plan slide instead of sitting on the Square 3 slide — see Deck Structure below.

Flagging Any "Not Enough Information" Moment (Yellow Flag)

This is a blanket rule, not something scoped only to Square slides: anywhere in the deck where the source material doesn't have enough to state something as a confirmed fact — a whole slide, a single bullet, a missing date, an unscored metric, a budget figure that was never resolved — that spot gets visually flagged yellow so the CMO can scan the deck and immediately see every place that needs their attention before it goes to the client. Don't rely on the wording alone ("timing not specified," "not yet addressed") to carry that signal — the CMO should be able to spot every gap at a glance without reading every line.

Where this shows up, beyond the Square slides themselves:

1. Any unaddressed Square — trailing or mid-sequence, no exceptions. Every Square that wasn't actually run gets its own explicit slide now, regardless of whether it's a trailing gap (nothing run comes later than it — the engagement just hasn't gotten there yet) or a mid-sequence gap (a later Square was run, so this one was actively passed over). This replaces any prior instinct to skip trailing gaps silently — don't skip them anymore. Still classify and word them differently (see below), but both get a slide, and both get the yellow flag.
2. Any addressed Square that's thin on material — i.e. the Square was actually run/discussed, but there wasn't enough in the source to fill in its framework. For Square 1 specifically, this means any of the four persona categories above (Pain points / Desired outcomes / Barriers to investment / Purchase criteria) that the source didn't cover in enough depth gets its own yellow card on that persona's slide. For Square 2, a missing Storybrand brand story (if one wasn't actually produced) simply isn't included at all rather than flagged — it's not a required element the way the message list is. Elsewhere, e.g. a Zero Square/Growth Trajectory slide missing CAC/LTV numbers, or a Square 3 slide where a channel's Rented/Owned classification is genuinely unclear from source. The slide still gets built with whatever real content exists, but the flag signals to the CMO that this one needs follow-up before it's presented as settled.
3. The Zero Square & Growth Trajectory slide — any target/metric that's directional or partially defined rather than a clean, sourced number gets its own yellow card rather than sitting styled identically next to fully-confirmed numbers. (This slide now also carries the consolidated growth objectives that used to live on their own post-Square slide — see Deck Structure.)
4. The Implementation Plan slide — any Key Result or tactic with unspecified timing, an unassigned owner, or any other missing planning detail gets a yellow card/line, per the existing "timing not specified" guidance — extended to any missing planning detail, not just dates. The new "What's at Stake" line (see Deck Structure) follows the same rule: if the source doesn't actually support a specific stakes statement, flag it yellow rather than writing generic urgency language to fill the space.
5. Anything that would have gone on a standalone Open Decisions slide. This deck no longer has one (see Deck Structure) — a still-open item, disagreement, or unclear question now surfaces in speaker notes on whichever slide it's most relevant to, or lives only in the written marketing plan document when one exists this session. Speaker notes don't get a visual yellow fill (they're not rendered on the slide itself), so the honesty standard here is about wording, not color: state plainly in the notes that the item is unresolved, don't imply it was settled just because it isn't visually flagged.
6. The Pain slide and the closing We Solved This slide — if the source doesn't describe the client's before-state in enough detail to state it confidently, or if the "here's how we solved it" callback can't be tied to something concrete in the plan, flag it yellow rather than writing something vague-but-confident-sounding.

Visual treatment: fill the slide's content area (or, for a slide that's otherwise mostly complete, just the specific card/box/line holding the thin item) with a warm yellow background — #FFF3CD fill, #F0B429 border/accent bar, near-black (#0A0A0A) text. This is a deliberate exception to the "no secondary accent color" rule in references/house-design-system.md — yellow is reserved exclusively for this flagging purpose and should never appear anywhere else in the deck, so it stays unambiguous as a "needs attention" signal the moment the CMO sees it. This yellow is Claude's own addition, not part of the house design system itself — never let it bleed into other slides' styling.

Wording still matters even though the visual is now the same shape for both gap types:

• Trailing gap slide: short, calm — e.g. "Not yet addressed this sprint" / "Scheduled for a future session." Don't pad it with invented content; a title, the yellow fill, and one honest sentence is the whole slide.
• Mid-sequence gap slide: same yellow fill, but the sentence should reflect that it was actively passed over — e.g. "Not yet addressed this sprint — deferred from the original agenda."
• Thin-square slide: keep the real content in its normal (non-yellow) cards, and flag only the specific missing piece — e.g. a single yellow card reading "Barriers to investment — not discussed this sprint" sitting alongside a persona's other, normally-styled cards.
• Never merge a trailing gap and a mid-sequence gap onto the same slide, and never let the flag wording imply a decision was made when it wasn't — the Core Rule still governs: no invented substance, ever, flag or no flag.

Keep the Agency's Own Commercials Out of the Early Slides

The Zero Square & Growth Trajectory slide (and, by the same logic, any early Square slide) exists to spell out the client's numbers — their growth goal, their implied marketing budget ceiling, their CAC/LTV math, their consolidated growth objectives — not to pitch the agency's own fee. Even when a specific agency retainer or execution-fee number was genuinely resolved during the sprint (a real, sourced fact), presenting it this early in the deck reads like a sales pitch before the plan itself has even been laid out, which undercuts the "here's your problem, here's the plan" narrative this deck is built around.

• On the Zero Square & Growth Trajectory slide: state the budget ceiling/implied annual marketing spend and growth targets in the client's own terms (CAC, ACV, revenue goals, etc.) without folding in a line item for what the client pays the agency specifically.
• If an agency retainer/fee figure was resolved in the sprint, it's still real and can still appear later in the deck — on the Implementation Plan slide if it's material to how the tactics get funded/executed, or on the closing We Solved This + Ongoing Services slide (see Deck Structure) where a commercial conversation is expected and appropriate. Don't drop it into the Zero Square & Growth Trajectory slide just because that's where the budget math lives.
• Square 3's sequencing/pacing decisions (see Square 2 and Square 3 guidance above) follow the same logic when they land on the Implementation Plan slide — they're execution decisions, not a plain channel list, so they belong where execution gets planned, not on the early channel-list slide.
• This is a sequencing rule, not a suppression rule — the number doesn't get hidden or invented away, it just moves to a point in the deck where a commercial conversation is contextually appropriate rather than premature.

Deck Structure

Produce slides in this order:

1. Title / The Challenge. The deck title, client name, and date, but led by a one-line hook naming the client's actual situation before this engagement — not a bare title slide. Pull this from source per the "before this engagement" reading guidance above; it should read like the opening beat of a story, not a logistics header. Sprint logistics that used to occupy their own slide (exact date, session format, who facilitated, who was in the room) do not get a slide anymore — fold anything worth keeping into this slide's speaker notes instead, so a presenter still has it handy without putting it in front of the client.
2. The Pain. A dedicated slide laying out what the client was actually dealing with going into this sprint — the problems, costs, risks, or frustrations that led them to engage the agency. This replaces what used to be a "the journey / what actually happened day to day" slide — the client doesn't need a recap of which Squares were run in what order; they need to see their own problem stated back to them clearly, so the plan that follows reads as the answer to it. Source this the same way as the opening hook (see Inputs) — real, specific pain points from the transcript, not generic industry pain. If the source genuinely doesn't describe the client's own before-state in any detail, say so plainly on this slide (yellow-flagged, per the rules above) rather than inventing a plausible-sounding problem statement.
3. Zero Square & Growth Trajectory. One slide (or a tight two-slide pair, if the numbers don't fit cleanly on one) carrying the client's own top-line math together, rather than splitting the budget math from the growth objectives the way earlier versions of this deck did: the growth goal, the implied marketing budget ceiling, and the consolidated growth objectives/targets (revenue goal, interim goals, ACV, CAC, and any other measurable target stated across the sprint) — all stated in the client's own terms (see "Keep the Agency's Own Commercials Out of the Early Slides" above — no agency retainer/fee line item here). This absorbs what used to be a separate "Objectives" slide presented after all nine Squares — pulling those numbers forward to sit with the rest of the top-line math is a deliberate structural change based on direct feedback that a standalone post-Square Objectives slide read as disconnected from the trajectory numbers it was built from. This slide gets the same conditional treatment as the numbered Squares in step 4 below: if the underlying Zero Square work wasn't run, it still gets a slide now (per the yellow flag rules above), marked with the flag rather than left out entirely.
4. One slide per Square, 1 through 9, no exceptions now — except Square 1, which becomes one slide per persona (see Square 1 guidance above), and Square 2 and Square 3, which reduce to a plain list as their primary content (see Square 2 and Square 3 guidance above). For every Square that wasn't run, classify it before deciding how to word it — don't handle unrun Squares as one lump group:
   • Determine each unrun Square's position independently. Find the highest-numbered Square that was actually run. Any unrun Square that comes after every run Square (nothing run comes later than it) is a trailing gap — the sprint just hasn't gotten there yet. Any unrun Square that has at least one run Square after it in the numbering is a mid-sequence gap — it was actively passed over.
   • Both kinds now get their own explicit slide, styled with the yellow flag treatment (see above) and worded per the distinction above — trailing gaps read as "not reached yet," mid-sequence gaps read as "deferred."
   • Never merge a mid-sequence gap and a trailing gap onto the same slide, even if that's convenient. A single "Squares 4-5 & 8-9: Not Yet Addressed" slide flattens a real gap the client actively noticed (4-5, passed over mid-sprint) into the same bucket as a gap that's simply next on the list (8-9, never reached yet) — it either overstates the trailing gap as a deliberate omission or understates the mid-sequence one as routine. If multiple Squares of the same kind are contiguous (e.g. 4 and 5 are both mid-sequence gaps back-to-back), one shared slide for that group is fine — the rule is about not mixing kinds, not about never grouping.
5. Implementation plan slide. Same derivation as the plan document's Implementation Plan section, kept just as lean: the goal, a couple of dated key results, a short tactics list under each. Add a short "What's at Stake" line or callout near the Company Goal / Marketing Goal boxes — the real stakes or urgency behind hitting (or missing) this goal, if the source actually describes any (a runway concern, competitive pressure, a deadline with real consequences, whatever fuel is genuinely there). Source this like any other Core content; if the source doesn't support a specific stakes statement, flag it yellow rather than writing generic "this matters a lot" language to fill the space. This slide is also where Square 3's sequencing/pacing decisions land (see above), and where an agency retainer/fee number is appropriate if material to how this plan gets executed (see above). Don't build a generic "Phase 1 / Phase 2 / Phase 3" graphic to make the plan look more sequenced than it is; flag any missing timing, owner, or other planning detail per the yellow flag rules above rather than inventing one.
6. Next Steps. Expand this beyond a single flat owner/action/timing table into a fuller sense of the current engagement getting underway: group the real next steps by phase or timeframe that the source actually supports (e.g. Immediate, Pre-Kickoff, Kickoff, Ongoing Cadence — use whatever phase breakdown the source's own next-steps material implies, don't force a rigid four-phase template onto a sprint that only has two real phases of next steps), and call out the kickoff date/meeting as its own prominent stat or callout rather than one more row in a table. The goal is a slide that reads as "here's what's already in motion," not a bare task list. (This is modeled on a fuller "current engagement" treatment from a past client deck that isn't available as a direct reference file in this environment — if such a reference deck is ever supplied, match its exact structure directly instead of this interpretation.)
7. We Solved This + Ongoing Services. The closing slide, and it must explicitly reference back to The Pain (slide 2) by name — e.g. if slide 2 named "21,000 dormant leads and no nurture system," this slide's opening line should read as the resolution to exactly that, not a generic "thanks for the opportunity" close. One or two sentences tying the plan just presented back to that original pain, then the agency's own ongoing-services tiers presented as the continued-engagement option. Source the tier names, what each includes, pricing, and any "ideal for" notes from the agency's own rate card — this skill doesn't ship tier data, and don't invent tiers or prices. Present them as a row of tier cards (one card per tier: name, price or price basis, a one-line "ideal for", and a short inclusions list). Always include the ongoing-services tiers here regardless of whether this specific sprint touched Square 8 — they're the agency's standing offer, not something that needs its own transcript citation, so this slide doesn't get yellow-flagged for using them. This is the other appropriate place (alongside Implementation Plan) for a resolved agency retainer/fee figure, since a commercial conversation is expected here. The "here's how we solved it" line, however, must trace to what the plan actually contains, same as any other Core content — flag it per the yellow flag rules above if it can't be tied to something concrete.

This deck no longer includes a standalone Open Decisions slide. Direct client-side feedback was that a dedicated "here's what's still unresolved" slide isn't necessary in front of the client — the client doesn't need a recap slide of everything still up in the air; they need the plan, with gaps visible where they naturally occur. Open, disputed, or unresolved items still matter and still need to be tracked — they just move to speaker notes on whichever slide they're most relevant to (a still-open persona question in that persona's speaker notes, a disputed budget figure in the Implementation Plan speaker notes, a rejected-but-maybe-worth-revisiting segment in that Square's speaker notes), and/or live in the written marketing plan document's own Open Decisions section when sprint-marketing-plan-builder has been run alongside this deck. Removing the slide is not license to drop the honesty standard — an open item is still stated as open wherever it ends up, just not staged as its own client-facing slide.

Process

1. Confirm source and scope. Which source material, which Squares were actually run, and whether a sprint-marketing-plan-builder extraction already exists for this sprint that can be reused.
2. Draft slide by slide, one arc at a time — the opening (Title/Challenge, then The Pain) first, then the plan arc (Zero Square & Growth Trajectory through Squares 1-9 — remembering Square 1 expands to one slide per persona, and Squares 2-3 reduce to plain lists — plus the Implementation Plan), then the closing arc (Next steps, We Solved This + Ongoing Services). Don't jump between arcs; it's easy to lose track of which Squares still need a slide if you're bouncing around.
3. QA pass before presenting. Eight checks, all required: does every Core content bullet on every slide trace to source; does every open or contested item get captured in speaker notes (or the plan document) rather than silently dropped now that there's no standalone Open Decisions slide; does every unrun or thin Square actually carry the yellow flag treatment; does the closing slide explicitly and specifically reference the opening Pain slide rather than closing generically; does the Zero Square & Growth Trajectory slide stay free of any agency retainer/fee line item while still carrying the consolidated growth objectives; does every Square 1 persona get its own slide structured on the four Buyer Persona Institute categories rather than a combined slide; do the Square 2 and Square 3 slides read as plain lists (messages + Storybrand story; channels split Rented/Owned) with process/sequencing detail moved out to speaker notes or the Implementation Plan; and — the broadest check — scan every single slide (not just Square slides) for any bullet, date, number, or claim that isn't fully sourced, and confirm each one carries the yellow flag rather than reading as confidently stated.

Output Format

The deliverable is a finished .pptx file, not a text outline handed back in the chat. This skill is meant to be run by any Fractional CMO on any client's sprint material and produce something ready to present — don't stop at drafting the slide-by-slide text.

Once the content is drafted and QA'd against the Core Rule:

1. Consult the pptx skill for the actual deck-building mechanics (pptxgenjs gotchas, layout, color palette, speaker notes via addNotes(), required validation and visual QA render). This skill owns the words; the pptx skill owns turning them into a real deck. Read it, then use it.

2. Keep the design clean and simple, matching the simplicity principle above — a clear, consistent look (one palette, one motif) rather than an elaborate one. This is a working recap deck, not a brand pitch — legible and calm beats dense and decorated. Still give every slide at least one non-text visual element per the pptx skill's design guidance (a stat callout, a simple icon, a card) — plain title-and-bullets slides are hard to present from and easy to skim past.

   Default to the house style in references/house-design-system.md (colors, fonts, card patterns, divider-slide layout, and the logo/motif assets you supply in assets/) as a starting point. Specifics worth holding onto from that reference: numbered section dividers are full-bleed primary blue with a near-black numeral (not a black background); cover-page and plain transitional-divider titles use heavy Montserrat rather than the serif (Libre Baskerville is for in-page slide headlines only); and there are several valid content patterns beyond a single white-bordered card — plain stat-number-and-label pairs with no card at all, flat borderless light-gray card grids, and a compact Before/During/After 3x3 grid that maps directly to Squares 1-9. Layer the yellow flag rules above on top of whichever palette is used as the one deliberate exception. If the client has their own brand guide or the user hands you a different reference deck to match, that always overrides this default (keep the yellow flag convention regardless, since it's a gap-visibility signal rather than a brand element); this is a fallback, not a mandate to brand every deck with the agency's own identity.

3. Name the file after the client: {Client Name}_Sprint_Deck.pptx. If the client name isn't obvious from the source, ask rather than guessing or leaving a placeholder.

4. Run the pptx skill's required validation and visual QA (render to images, check every slide for overflow or overlap) before calling this done — including confirming the yellow flag slides/cards render with the correct fill and aren't mistaken for a design error.

If a user explicitly asks for a text outline instead of a file, that's fine — but the default assumption, especially when this skill is being used hands-off by someone who wasn't in the room for the sprint, is a finished deck they can open and present as-is.

Non-negotiables

• Never write a slide's Core content from a generic sprint-recap template — every Square slide must reflect what THIS client's sprint actually decided, not a stock description of what that Square usually covers.
• Never invent a Core content bullet or a factual speaker-note claim without a source; the speaker-notes exception covers delivery guidance only, never substance.
• Open with the client's own, sourced pain — never a generic industry-pain statement, and never a bare logistics/title slide. Sprint logistics (date, format, facilitators, attendees) live in speaker notes now, not their own slide.
• Give every persona from Square 1 its own slide, structured on the four Buyer Persona Institute categories (Pain points, Desired outcomes, Barriers to investment, Purchase criteria/questions) — never combine multiple personas onto one slide, and keep methodology/rejected-alternative notes off the slide itself (speaker notes, or the plan document, instead).
• Keep Square 2 and Square 3 as plain lists — messages (+ Storybrand story, if one exists) for Square 2; channels split Rented/Owned for Square 3 — with process/status/sequencing detail moved to speaker notes or the Implementation Plan rather than mixed into the list.
• Carry the consolidated growth objectives on the Zero Square & Growth Trajectory slide, not on a separate post-Square Objectives slide.
• Do not include a standalone Open Decisions slide in the deck — route open/disputed items to speaker notes on the relevant slide, or to the written marketing plan document, and still state them plainly as unresolved wherever they land.
• Close by explicitly tying back to the opening Pain slide before presenting the Ongoing Services tiers — a generic "thanks" close doesn't satisfy this.
• Keep the Zero Square & Growth Trajectory slide free of any agency retainer/fee line item — state the client's own budget ceiling/math and growth targets there, and save the agency's own commercial number (if one was resolved) for the Implementation Plan or the closing slide.
• Give every unaddressed Square — trailing or mid-sequence — its own explicit, yellow-flagged slide. Never skip a Square silently anymore, and never template filler content to avoid the gap; the flag itself, plus one honest sentence, is the whole slide.
• Flag any "not enough information" moment anywhere in the deck with the same yellow treatment — not just Square slides. This explicitly includes the Zero Square & Growth Trajectory slide, the Implementation Plan slide (including its "What's at Stake" line), the Pain/We-Solved-This slides, and any of the four persona categories on a Square 1 slide, scoped to just the specific card/box/line that's thin rather than the whole slide if the rest is genuinely complete.
• Reserve yellow exclusively for this flagging purpose — never use it decoratively or for anything else in the deck.
• If a same-session sprint-marketing-plan-builder extraction exists, reuse it rather than re-deriving it, to keep both deliverables consistent with each other. If it doesn't exist — because that skill isn't available or wasn't run — derive everything directly; this skill must never treat the plan document as a hard prerequisite.