Open Design To Open Slide logo

Open Design To Open Slide

Community
dracohu2025-cloud
open-design-to-open-slide

Use when mining Open Design visual systems and converting them into Open Slide React template kits or standalone 20-page template albums, with build, deployment, screenshot QA, and provenance rules.

Overview

Publisherdracohu2025-cloud
Repositorydraco-skills-collection
Skill nameopen-design-to-open-slide
Stars
228
Forks
37
Bundled files
64
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 64 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by dracohu2025-cloud on GitHub. Read the source before you install it.

Installation

Install the Open Design To Open Slide AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/dracohu2025-cloud/draco-skills-collection/tree/main/Ark%20Agent%20Plan%20Version/open-design-to-open-slide
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Use it in TypingMind

Enable Open Design To Open Slide in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Open Design To Open Slide on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Open Design To Open Slide is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Open Design → Open Slide Template Production

Overview

This skill captures a practical workflow: use Open Design as a visual template corpus, then rebuild the result as Open Slide React slides.

Core rule: Open Design is a source of design language, not a runtime dependency. Mine its palettes, typography, page recipes, examples, and component rhythm; implement the output as Open Slide Page[] with React primitives.

This skill also bundles a ready-to-copy asset pack at templates/open-slide-template-pack/: 49 decks, 867 pages = 11 Open-Slide official/starter decks + 38 Open-Design-derived suites.

Quality baseline:

  • Treat original Open-Slide official examples as the stronger visual baseline.
  • Treat OD-derived suites as supplemental style expansion, not the flagship standard.
  • When improving OD suites, compare them against official Open-Slide examples before shipping.

Open-Slide Official Baseline Inventory

The playground includes 11 official/starter Open-Slide decks. Use these as the first reference set when judging layout, motion, hierarchy, and polish.

SlugSourcePagesRole
getting-startedCLI starter template13Starter / baseline primitives
open-slide-launchOfficial demo7Product launch narrative
open-slide-anatomyOfficial demo16Deep anatomy explainer
vercel-ai-sdkOfficial demo8Technical product explainer
ssh-explainedOfficial demo10Concept teaching deck
material-design-2014Official demo7Design history / visual storytelling
claude-code-introOfficial demo9Product intro / workflow
harness-engineeringOfficial demo8Engineering/product narrative
llm-fundamentalsOfficial demo12Knowledge explainer
nextjs-ppr-cacheOfficial demo8Short technical comparison
raycast-apiOfficial demo9API/product explainer

Official/starter total: 11 decks, 107 pages.

Bundled template pack inventory: 49 decks, 867 pages = 11 official/starter decks + 38 standalone OD suites.

Local playground inventory may include extra kit/demo decks, but the public skill pack intentionally ships only the clean 49-deck template set.

When to Use

Use this skill when you need to:

  • Convert Open Design examples or design systems into Open Slide decks.
  • Create an OD-derived template kit, suite, album, or gallery.
  • Expand a small template suite into a reusable 20-page presentation pack.
  • Build public-facing React slide templates with screenshot QA.
  • Preserve design provenance while avoiding runtime coupling.

Do not use it for:

  • Importing Open Design daemon, web app, agent adapter, or navigation runtime.
  • Copying raw single-file HTML decks directly into Open Slide.
  • Producing .pptx files.
  • Shipping public repos without credential and path audits.

Architecture Decision

LayerSourceKeepReject
Visual languageOpen Designcolor, typography, spacing, cards, diagrams, densitylogos, brand lockups, vendor identity
Page recipesOpen Design examplescover, agenda, metrics, diagram, timeline, quote, matrixraw HTML navigation scripts
RuntimeOpen SlideReact Page[], TSX primitives, static buildOpen Design daemon/runtime
QAHeadless browserscreenshots, hash checks, contact sheetstrusting build success only

Standard 20-Page Suite Structure

For OD-derived template suites, use this stable 20-page shape:

  1. Cover
  2. Agenda
  3. Problem / Context
  4. Framework
  5. Content
  6. Metrics / Data
  7. Timeline / Roadmap
  8. Diagram / Architecture
  9. Closing / CTA
  10. Section Divider
  11. Quote / Key Insight
  12. Comparison
  13. Process / Workflow
  14. Matrix / 2×2
  15. Table / Spec
  16. Case Study
  17. Checklist
  18. Risks / Tradeoffs
  19. FAQ / Appendix
  20. Thank You / Contact

Important learned preference:

  • Keep the first 9 pages visually stable.
  • Add coverage by appending pages 10–20.
  • Do not mass-rewrite many suites into heavily scene-specific layouts unless screenshots prove the result is better.

Bundled 49-Deck Asset Pack

The skill ships a complete Open Slide template pack:

text
open-design-to-open-slide/templates/open-slide-template-pack/
  manifest.json
  README.md
  slides/
    getting-started/index.tsx
    open-slide-launch/index.tsx
    ...
    od-xhs-post-suite/index.tsx

Install into any Open Slide project:

bash
bash open-design-to-open-slide/scripts/install-template-pack.sh /path/to/open-slide-project

Install only OD-derived suites:

bash
bash open-design-to-open-slide/scripts/install-template-pack.sh /path/to/open-slide-project --od

Install only official/starter decks:

bash
bash open-design-to-open-slide/scripts/install-template-pack.sh /path/to/open-slide-project --official

Overwrite existing same-slug decks only when explicit:

bash
bash open-design-to-open-slide/scripts/install-template-pack.sh /path/to/open-slide-project --overwrite

Inspect the bundled manifest:

bash
python3 open-design-to-open-slide/scripts/list-template-pack.py

Pack rules:

  • Keep slides as plain Open Slide React source under slides/<slug>/index.tsx.
  • Do not include Open Design runtime code.
  • Do not include local deployment output, screenshots, caches, or credentials.
  • manifest.json is the count source of truth for consumers.

Workflow: Create a New OD-Derived Suite

  1. Pick a slug:
bash
mkdir -p slides/od-<name>-suite
  1. Inspect Open Design source candidates:
bash
# examples only; adapt paths to your local Open Design clone
ls skills/
ls templates/
ls design-systems/
  1. Extract only:
  • palette
  • typography
  • component shape
  • visual rhythm
  • page archetypes
  • provenance path
  1. Implement in Open Slide TSX:
tsx
import type { CSSProperties, ReactNode } from 'react';
import type { DesignSystem, Page, SlideMeta } from '@open-slide/core';

export const design: DesignSystem = { /* tokens */ };
export const meta: SlideMeta = {
  title: 'OD <Name> Suite',
  description: '<Name> OD-derived 20-page Open Slide template suite.'
};
export const notes = [
  'Source: Open Design ...',
  'Runtime: Open Slide only'
];

const Cover: Page = () => /* ... */;
// ... 20 pages total
export default [
  Cover, Agenda, Problem, Framework, Content,
  Metrics, Timeline, Diagram, Closing,
  SectionDivider, QuoteInsight, Comparison, ProcessWorkflow,
  Matrix2x2, TableSpec, CaseStudy, ChecklistPage,
  RisksTradeoffs, FAQAppendix, ThankYou
];
  1. Keep design helpers local and simple:
tsx
const rgba = (h: string, a: number) => {
  const x = h.replace('#', '');
  const n = parseInt(x.length === 3 ? x.split('').map(c => c + c).join('') : x, 16);
  return `rgba(${(n >> 16) & 255}, ${(n >> 8) & 255}, ${n & 255}, ${a})`;
};

const isDarkBg = (h: string) => {
  const x = h.replace('#', '');
  const n = parseInt(x.length === 3 ? x.split('').map(c => c + c).join('') : x, 16);
  const r = (n >> 16) & 255;
  const g = (n >> 8) & 255;
  const b = n & 255;
  return (0.2126 * r + 0.7152 * g + 0.0722 * b) < 96;
};

const dark = isDarkBg(t.bg);

dark must be global if appended pages use it. Missing global dark can build successfully but render blank at runtime. Nasty little trap.

Workflow: Expand a 9-Page Suite to 20 Pages

  1. Count pages:
bash
python3 - <<'PY'
from pathlib import Path
base = Path('slides')
for p in sorted(base.glob('od-*-suite/index.tsx')):
    s = p.read_text()
    pages = s.split('export default [')[-1].split(']')[0].count(',') + 1
    print(p.parent.name, pages)
PY
  1. Append pages 10–20 from the standard recipe list.
  2. Update copy such as 9 slides → 20 pages.
  3. Update metrics such as 9 → 20 and slides → pages.
  4. Ensure appended pages only reference globally available helpers.
  5. Run build and screenshot QA.

Build and Deploy

Typical Open Slide build:

bash
pnpm install
pnpm build

For a static deployment, copy the build output to your web root. The exact path depends on your server.

Screenshot QA

For important work, screenshot every page and build contact sheets.

Minimum QA checks:

  • all expected screenshots exist
  • public route or local preview loads
  • DOM is not empty
  • no near-blank pages
  • screenshot hashes are unique unless duplication is intentional
  • cover sheet and full contact sheet are generated

Representative command pattern:

bash
CHROME=${CHROME:-chromium}
OUT=/tmp/od-suite-shots
mkdir -p "$OUT"
for slug in $(find slides -maxdepth 2 -path 'slides/od-*-suite/index.tsx' -printf '%h\n' | xargs -n1 basename | sort); do
  mkdir -p "$OUT/$slug"
  for p in $(seq 1 20); do
    "$CHROME" --headless=new --no-sandbox --disable-gpu --hide-scrollbars \
      --virtual-time-budget=6500 --window-size=1920,1080 \
      --screenshot="$OUT/$slug/p${p}.png" \
      "http://127.0.0.1:5173/s/$slug?p=$p"
  done
  echo "captured $slug"
done

Use Python/Pillow to compute hashes, detect near-blank images via image variance, and generate contact sheets.

Visual Rules

  1. Design for 16:9 fixed canvas.
  2. One idea per page.
  3. Use strong hierarchy and cards; avoid bullet dumps.
  4. Use Open Design for rhythm, not runtime.
  5. Keep provenance in notes or comments.
  6. Strip obvious third-party logos and brand marks.
  7. Avoid remote assets unless there is a strong reason.
  8. Generate screenshots and contact sheets before declaring success.

Common Pitfalls

  1. Copying Open Design HTML directly. This creates a hidden second runtime. Port the design into React primitives instead.

  2. Missing global variables in appended pages. Pages 10–20 often reference dark, rgba, font, t, Card, Shell, Tag, H, and P. If any are scoped inside old functions, the deck may build but render blank.

  3. Trusting route 200. SPA route 200 only proves index.html was served. Use browser DOM/screenshot checks to catch white screens.

  4. Over-fitting each suite into scene-specific structure. Preserve visual coherence first. Expand coverage by appending reusable pages.

  5. Huge screenshot jobs with no progress. Use background execution and print progress per suite.

  6. Dismissing duplicate hashes. Duplicate hashes may mean actual white screens. Check variance and DOM.

  7. Leaking local paths or internal URLs in public docs. Public skills should use placeholders and environment variables.

Verification Checklist

  • Every suite exports the intended number of pages.
  • pnpm build passes.
  • Preview route opens in a browser.
  • Screenshots cover every suite/page.
  • Screenshot hash count equals screenshot count unless duplication is intentional.
  • No near-blank pages by variance check.
  • Contact sheet generated.
  • Public version contains no secrets, private IDs, or machine-specific absolute paths.
  • templates/open-slide-template-pack/manifest.json reports 49 decks.
  • scripts/install-template-pack.sh copies decks into a clean Open Slide project.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

and 4 more files.

Frequently asked questions

What does the Open Design To Open Slide AI skill do?

Use when mining Open Design visual systems and converting them into Open Slide React template kits or standalone 20-page template albums, with build, deployment, screenshot QA, and provenance rules.

Why use Open Design To Open Slide on TypingMind?

Because you install it once and use it with any model. Open Design To Open Slide is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Open Design To Open Slide in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/dracohu2025-cloud/draco-skills-collection/tree/main/Ark%20Agent%20Plan%20Version/open-design-to-open-slide. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Open Design To Open Slide?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Open Design To Open Slide?

As many as you like. As long as a model supports skills, you can use Open Design To Open Slide with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Open Design To Open Slide AI skill free?

Yes. It is published on GitHub by dracohu2025-cloud under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇