# Presentation Design > Generate a premium 6-slide presentation design board as one single composite image. Use this skill whenever the user asks for PPT design concepts, keynote-style slide mockups, a strategy deck visual, an editorial / sports / news / business analysis deck, a TV-friendly slide layout, or any "6-page slide board" / "design these slides" / "PPT 设计稿" / "幻灯片设计图" request. The skill plans the deck slide-by-slide, asks the user to pick light vs dark mode, selects layouts from a named layout library, has the user approve the plan, then generates the composite image and self-reflects on the rendered result. Do NOT trigger for text-only deck outlines (the user wants writing, not visuals), single-slide design (this skill always produces 6), or live editable PPTX/Keynote files (this skill produces a rendered image). - Skill: `agentara/presentation-design` (Agent Skill, multi-file: 3 files) - Install (CLI): `npx skillmds@latest add agentara/presentation-design` - Raw SKILL.md: https://api.skillmd.com/api/skills/agentara/presentation-design/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Docs & Writing - Author: agentara (https://skillmd.com/u/agentara) - Updated: 2026-09-10 - Page: https://skillmd.com/skills/agentara/presentation-design --- # Presentation Design Create a **premium 6-slide presentation design board** as one single composite image — a board view that lets the user see the whole deck at a glance, TV-ready and cinematic. Primary use cases: PPT design concepts, keynote-style slide mockups, strategy deck visuals, event / news / sports / business analysis decks, editorial data stories, executive presentations. ## Image Generation Requirement The final board must be generated by an **image-generation model** (use the built-in `image_gen` tool by default, or an approved project image-generation workflow). Do not create the board with code, SVG, HTML/CSS, canvas, diagrams, placeholder drawings, or deterministic shape scripts. Those methods may be used only for non-visual file operations or optional post-processing after the model-generated image exists. Save the model-generated raster output to `presentations/{name}.png`. If the model output cannot be saved directly, locate the generated raster asset and move/copy it into the required path; do not recreate it programmatically. ## Output Contract One single image containing all 6 slides: ```text ┌──────────────┬──────────────┐ │ Slide 01 │ Slide 02 │ ├──────────────┼──────────────┤ │ Slide 03 │ Slide 04 │ ├──────────────┼──────────────┤ │ Slide 05 │ Slide 06 │ └──────────────┴──────────────┘ ``` - **6 slides total** - **2 columns × 3 rows** - **each slide visually reads as 16:9** - **overall composite board is 3:2** - single composite PNG ## Output Location ```text presentations/ {name}.plan.md # the approved deck plan (spec) {name}.png # the final composite image ``` Use lowercase, hyphenated, ASCII-safe names. For revisions, suffix with `-v2`, `-final`, `-alt-a`. Do not overwrite existing files unless the user asks for replacement. ## Inputs to Extract Before designing, identify: - Topic / subject of the deck. - Audience (executives, fans, general public, B2B buyers, students, etc.). - Tone (cinematic, analytical, celebratory, urgent, premium, playful). - Brand colors or required accents, if any. - Required language for slide text (English, Chinese, bilingual, etc.). - Any specific data, charts, maps, portraits, products, or symbolic visuals that must appear. - Output name, if specified. If the user does not specify language, use the user's local language. ## Workflow 1. **Extract inputs** from the user's brief (see *Inputs to Extract*). 2. **Ask the user for style mode: `light` or `dark`** — always ask via `AskUserQuestion` (do not silently infer). Offer `dark cinematic` as the default suggestion when the topic is ambiguous. 3. **Load `references/style-modes.md`** and pick the most appropriate sub-mode (e.g. `dark sports broadcast` for an NBA recap, `light minimal keynote` for a B2B SaaS pitch). Briefly state the chosen palette and typography direction to the user. 4. **Load `references/slide-layouts.md`** and select **6 layouts** that together form a coherent narrative arc. Mix layouts — do not repeat the same one across all 6. Always use `hero-title` for slide 01 and `closer` for slide 06 unless the topic demands otherwise. State the chosen layout sequence. 5. **Write the deck plan** (mandatory) to `presentations/{name}.plan.md`. For each of the 6 slides include: - `slide_no` — `01`–`06`. - `layout` — one of the names from the layout library. - `title` — the headline that will appear on the slide. - `subtitle` — optional one-line framing. - `content` — 3–6 short bullets / key points in slide-native language (no essay paragraphs). - `visual` — description of the hero image, chart, map, portrait, or symbolic visual that occupies the tile, plus how it integrates with the text. - `accent_use` — where the accent color appears on this slide (title, divider, chart highlight, etc.). Show the plan to the user and **wait for confirmation or revisions**. If the user revises, update `presentations/{name}.plan.md` and re-confirm before proceeding. 6. **After the user approves the plan, generate the first composite image** with `image_gen`. Build the prompt by composing: - Output contract (3:2 board, 2×3 grid of 16:9 tiles, single PNG). - Selected style mode + palette + typography direction from `references/style-modes.md`. - For each of the 6 slides: layout name + title + content bullets + visual description + accent usage from the approved plan. - Content density rules and board-level composition rules (see below). Save the result to `presentations/{name}.png`. 7. **Self-reflect on the rendered image** (mandatory). Inspect the saved PNG and run both passes — *Pass 1: Structure & Story* and *Pass 2: Visual Polish* — against the actual rendered result, not just the plan. Check specifically for: - Grid is 2×3 and each tile reads as 16:9. - Slide titles are present, legible, and match the plan. - Accent color is used consistently across all 6 slides. - Image treatment looks integrated, not pasted-on stock-photo. - Layouts on each tile match the assigned layout names. - No redundant or off-topic slides. - Charts, maps, numbers, and quotes are readable on a TV-sized screen. 8. **Decide: ship or regenerate.** If self-reflection finds material issues, regenerate — either re-render the full board with a tightened prompt, or do a targeted fix (re-prompt focused on the broken slide(s), instructing the model to preserve the rest of the board). Cap automatic regeneration at **2 retries** before surfacing remaining issues to the user. Save the final accepted image to `presentations/{name}.png` (overwrite prior drafts); optionally keep `presentations/{name}-v1.png` etc. for debugging. 9. **Respond to the user** with: the saved image path, the deck plan path, the style mode chosen, the layout sequence, and a short note on anything the self-reflection could not fully resolve. ## Content Density Rules Each slide should be rich but not cluttered. Preferred density per slide: - 1 main title. - 1 short subtitle or framing line (optional). - 3–6 short content points. - 1 dominant visual system (hero image, chart, map, portrait, product, or symbolic scene). - Optional chart, metric card, or timeline element. Avoid: essay-like writing, too many bullet points, overexplaining, raw article summaries, tiny unreadable text, excessive icons, inconsistent information density. Use slide-native language: short, sharp, visual, structured, easy to scan. ## Board-Level Composition Rules - The full image is **3:2**. - The 6 slides are arranged in **2 columns × 3 rows**. - Slides have clean gutters or subtle borders between tiles. - Consistent margins across the board. - All 6 tiles share the same style mode, accent system, and typography — the deck must feel unified at a glance. - Accent color usage must be consistent across all 6 slides. - Each slide should have a clear focal point, strong title hierarchy, enough breathing room, integrated visuals, readable text, and the premium presentation feel defined in *Quality Bar*. ## Self-Reflection Loop Always run both passes before declaring the board final. ### Pass 1 — Structure & Story Check: - Is the 6-slide story coherent? - Does each slide have a clear role? - Is the order logical? - Is anything missing? - Is any slide redundant? - Is the content too vague or too crowded? Improve: slide sequence, narrative clarity, content hierarchy, factual compression, section naming. ### Pass 2 — Visual Polish Check: - Does the board look premium? - Is the chosen mode (dark or light) elegant, not muddy or flat? - Is the design suitable for TV display? - Are the images naturally embedded? - Are there repeated or boring layouts? - Is the contrast strong enough? - Do all 6 slides feel like one deck? Improve: contrast, spacing, typography, color balance, hero image integration, chart readability, cinematic depth, consistency. ## Quality Bar The final design **should** feel like: - a premium conference deck - a high-end sports broadcast graphic package - a cinematic strategy presentation - an editorial data story - a TV-ready executive presentation The final design **must not** feel like: - a random collage - six unrelated posters - a comic page - a low-effort infographic - a plain document - a cluttered dashboard - a generic template ## References Load these on-demand during the workflow — they are not read by default to keep this file lean. - [`references/slide-layouts.md`](references/slide-layouts.md) — named layout library. Load at workflow step 4 when choosing the 6-slide composition. - [`references/style-modes.md`](references/style-modes.md) — light / dark sub-modes with concrete palettes and typography direction. Load at workflow step 3 after the user picks light or dark.