open-kimi-ppt-skill/skills/open-kimi-ppt/reference/design_system/promotion/pine-soot-pictorial/design.md

Pine Soot Pictorial · Promotion STYLE DESIGN SYSTEM

PART A — Category Baseline (General Minimum)

An emotion-led yet fully auditable brand presentation, benchmarked against annual reviews from top-tier agencies, campaign retrospectives, and destination/product launch decks. The deck must land one memorable creative idea on every slide while keeping every claim verifiable—expressive in form, disciplined with numbers.

Content and layout:

1. Creative-led structure: one unifying annual/campaign proposition → audience and market evidence → big-idea development (waves/chapters) → channel and content system → results with full measurement context → direction for the coming year.
2. Make big numbers the protagonists: give a small set of headline metrics full-slide or oversized emphasis; arrange supporting evidence around them in compact modules.
3. Build rhythm at medium density: alternate evidence-dense slides with spacious statement slides; the deck should rise and fall like a story, not fill every surface like a report.

Visual style:

1. Aesthetic: confident, contemporary, and brand-first; use photographic or illustrative texture only when it is authentic to the source style—never pad the deck with generic stock imagery.
2. Typography: one expressive display face carries the major propositions, as defined by the style signature, while one highly legible workhorse face handles body copy and data; set large numerals in the display face; keep case and tracking rules consistent throughout.
3. Color: give each deck one brand-color family—one primary color with lighter and darker steps—plus one contrasting accent reserved for CTAs and highlighted data; let the style signature determine the background; no rainbow charts.
4. Graphics: reduce charts to the elements required by the narrative—few series, direct labels; timelines, funnels, and audience bars are signature graphic forms; icons must follow the style’s established line or fill language, with no multicolor clip art.

Structure and discipline (distilled from the work we recognize as benchmark quality):

1. One big idea per slide: the slide title states the campaign proposition or chapter message, and every module on the slide serves that one idea.
2. Campaign evidence must include full measurement context: impressions, completion rate, engagement rate, and conversion must always state platform, period, and sample scope; vanity metrics without context are a defect.
3. Timeline slides are first-class citizens: every wave, season, or phase has a named cadence and phase-specific metrics; a campaign deck without a rhythm slide is incomplete.
4. Audience slides use funnels, tiers, or audience bars—demographics must be tied to behavior and value, never shown as bare percentages.
5. Results slides distinguish outputs (impressions, content volume) from outcomes (search lift, new-customer share, sell-through rate); outcomes take visual precedence over outputs.
6. Case-study deep dives use fixed fields: context → action → numerical evidence → implication; parallel cases must use consistent, comparable fields.
7. Social/UGC evidence cites authentic order of magnitude and shows the structure of representative content; redraw all third-party platform screenshots.
8. The ending must establish direction: next year’s theme, total budget and allocation, and measurement framework—a closing slide that says only “thank you” is a defect.
9. Every number must include its unit, point in time, and source/measurement context; label estimates as internal illustrative assumptions.
10. Imagery and shape language: when permitted by the style signature, proactively source authentic imagery and replace generic rectangular blocks with expressive shapes—torn edges, collage, irregular crops, and signature forms; flat rectangles are a last resort, not the default.

Default baseline (distilled from the layout skill; PART B may override): one proposition per screen, with each slide communicating only one core message and no more than 3 continuous lines of body copy; visual tension must come from at least one clearly defined device—expansive negative space, strong contrast, a large field of the primary brand color, or a full-bleed product/person image—and remain consistent throughout; no table-heavy layouts, with data expressed first through oversized numerals, a single chart, or comparison bars; use one decorative language across the deck and never mix more than one system.

Title Read-Through Test: read only the slide titles from beginning to end; they should tell the complete story. If the sequence does not flow, the structure has not been established.

Density anchor: the specific minimums for each style—the number of evidence modules and charts per slide, plus annotation conventions—follow the 【Density Baseline】 reverse-engineered from that style’s source material; this baseline does not impose a universal number.

Global Prohibitions (AI Default Tropes · Mandatory)

1. Do not default to cards: unless the user explicitly requests them, never use rounded rectangles or rectangular cards to construct hierarchy or alignment—functional containers such as status indicators, assumption boxes, and quotation frames do not count as cards. Rules, negative space, type scale, and weight contrast are preferable.
2. Do not default to equal subdivisions: unless no other workable layout exists, do not default to three equal columns, four equal columns, or a two-by-two matrix—equal division must be justified by the content structure (three scenarios naturally support three columns; frameworks explicitly named in the body copy are exempt).
3. Do not use unmotivated AI color schemes: institutional palettes specified in the brief are exempt; the prohibition targets safe, familiar, unsupported defaults—blue-purple gradients, cyan-purple neon, rainbow flares, glass cards, and glowing borders.
4. Do not use elements that conflict with the overall style: no out-of-style treatments, such as rounded icons or rounded rectangles within a sharp-edged system.
5. Keep chart backgrounds transparent: use transparency or the slide background, except when reversing content out of a dark field.
6. Do not leave planned regions empty: the body-content area must not contain large blank zones; every planned region must be filled with real content, and failure to fill it is the most severe form of roughness.

PART B — Signature System (Distinctive Template Signature; Overrides PART A in Case of Conflict)

One-line style signature: a cinematic-pictorial editorial sensibility built on near-black grounds and full-slide cinematic photography, structured with white type, with cover red and magenta used as one-off bursts; large-scale imagery preserves breathing room within medium-high text density.

【Color Palette】Background #000101 (renders close to #000000): carries black fields and photographic negative space and may occupy 44%–56%. Primary text #FFFFFF: used for body copy, navigation, page numbers, and fine rules, never as a large filled field. Titles #FFFFFF: hierarchy comes only from size and weight. Primary accent #EF1722: the striped field at the cover’s lower right, occupying about one-fifth of the cover; prohibited in body slides. Secondary accent #FD1AEE: the narrow vertical stripe on the same cover, with roughly half the area of the primary accent. Chart series colors and positive/negative colors: not used; inherit PART A without adding new encodings. The interface layer on every slide is black and white only; the cover may add red and magenta, while photographic hues do not count. Iron rule: color comes only from photography or the cover stripes and never fills body-content containers.

【Layout Skeleton】Source facts: the canvas is 19×11 inches, at 19:11. Five micro anchors sit across the top at approximately x=2%, 25%, 48%, 70%, and 97%; the rightmost index is vertical, and the page number is fixed at the lower left. The main title begins at y≈31%, two-column body copy occupies y≈41%–74%, and the label group occupies y≈77%–91%; the reading path is navigation → title → explanation → labels. Photography and black field use a hard 43.7%/56.3% split that may be mirrored left to right; images meet the slide edge with square corners, and seams use black hairlines only. Cover only: approximately 59% of the left side is a full-bleed close-up, while the right 41% is black above and red/magenta vertical stripes below; the main title breaks into offset oversized words at the lower left. 16:9 migration: at 960×540, preserve the vertical hierarchy, two-column body, and square-edged division; assign the additional ≈2.9% width to the photographic crop or outer black field, place the division at 42%–44% (or mirror it), and never stretch the image or alter the vertical stack.

【Typography】Source typeface facts: sans-serif Inter (brought in via customFonts); body copy uses Inter, navigation uses Inter Medium, and titles and labels use Inter SemiBold. Body = 1x = 13 pt, body title ≈4.23x = 55 pt, oversized year numeral ≈2.77x = 36 pt, cover title ≈8.85x = 115 pt, and small label/footnote level ≈0.85x = 11 pt (there is no separate footnote style). Titles may run 2–3 lines, with line height around 0.6–0.7em and tracking tightened by approximately 4%–10%; compress title lines only and never overlap the body copy. Navigation and labels use ALL CAPS, body titles use sentence case, and four-digit indices are stacked vertically. For Chinese, choose a modern sans-serif with even stroke weight and keep to a single type family; continue using Inter for Latin letters and numerals. 16:9 migration: at 960×540, recommended sizes are 16–17 for body copy, 66–72 for titles, 124–144 for cover titles, and 13–15 for labels/navigation; preserve the original ratios and weights. These are not measurements taken from the source.

【Chart Language】Actual-use inventory: no data charts and no tables; only single photographs, horizontal storyboards, hard-edged image matrices, and attribute-label groups. ⭐Image matrix: compose 2–11 unequal-width frames into a square-cornered visual evidence wall; frame sizes may vary, but all boundaries must align. Use only to show multiple visual facets of the same subject, never to encode quantitative comparison. All other chart rules inherit PART A; no stable additions were observed in the source.

【Signature Components】Photography is a hard requirement: every body slide contains at least one cinematic photograph suitable for full-slide cropping, covering 44%–100% of the canvas; text sits only on a black field, a darkening layer, or natural negative space. Top micro-navigation + lower-left page number: each appears once per body slide, flush with the slide edge and outside the body grid; the leftmost anchor names only the section role and must not reproduce an institutional logo. Attribute-label group: one group appears in the lower portion of each body slide, containing 4–10 low-profile capsules with transparent fill and a fine #FFFFFF outline, set in Inter SemiBold ALL CAPS; arrange them across two lines according to content width, with no individual label wrapping. Cover red/magenta stripe field (occasional): use only once in the entire deck, confined to the lower-right quadrant, and never extend it into a decorative device for body slides.

【Prohibited】Do not use #EF1722, #FD1AEE, or any colored interface block on body slides; this preserves the impact of the cover burst. Do not use rounded photographs, floating cards, shadows, or wide white gutters; photography must be hard-cropped into a continuous composition. Do not place white type directly over highly textured or brightly lit photography; if there is no natural dark region, recrop first and use a localized black darkening layer only as the second choice. Do not use centered body-slide titles or symmetrical poster compositions; titles, body copy, and labels always begin on the same left edge. Do not substitute illustrations, 3D graphics, or multicolor icons for photographic evidence. Labels must not use solid fills, gradients, or multiline word breaks.

【Slide Types and Layouts】Black field—photography hard split｜Concept explanation + one set of visual evidence｜Approximately 44% photography and 56% black field, optionally mirrored, with two-column body copy in the black field｜3–5 independent blocks; photography may be 1 image or a 9–11-frame matrix｜Top navigation, label group, page number｜Unsuitable for high-dimensional data or long tables. Dual-photography split with type on dark regions｜One proposition requires a primary image and a supporting image｜Approximately 56% on one side for a type-bearing hero image with a dark region, and 44% on the other for 1–2 supporting images｜3–4 independent blocks｜Top navigation, label group, page number｜Unsuitable for photographs without clear negative space. Full-bleed photography with negative space｜A single strong visual supports a short analysis｜Photography bleeds across 100% of the slide; the title and two-column body copy occupy only approximately 42% on one side｜2–4 independent blocks｜Top navigation, label group, page number｜Unsuitable for parallel conclusions or dense parameters.

【Density Baseline】A typical body slide contains 1 main title, 2–4 explanatory blocks, 1 group of 4–10 labels, and 1–11 photographs; the chart count is always 0. Black-field split slides usually contain 3–4 text/label modules; image-matrix slides replace additional text with 9–11 frames; full-bleed photography slides retain only 2–3 text blocks. Source-language blocks contain approximately 120–400 characters; new drafts should target a language-equivalent concise length of 45–90 words/characters across 2–6 lines. Do not caption every image or annotate every point. Concentrate negative space between navigation and title, to the right of the title, within dark photographic regions, and below the label group; do not impose a universal module minimum across the deck—the more fully photography occupies the slide, the less text it should carry.