Presentation Extract Template logo

Presentation Extract Template

Organization
okou-ai
presentation-extract-template

Extract source style and editable HTML layouts from HTML decks (files, directories or URLs), PPTX, PPT, PDF or slide images, then save a reusable Custom Template.

Overview

Publisherokou-ai
Repositoryokou-skills
Skill namepresentation-extract-template
Stars
81
Forks
19
Bundled files
85
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.

  • 85 bundled files

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

  • Open source

    Published by okou-ai on GitHub. Read the source before you install it.

Installation

Install the Presentation Extract Template 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/okou-ai/okou-skills/tree/main/extract-template/presentation
  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.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/okou-ai/okou-skills.git /tmp/okou-skills
mkdir -p .claude/skills
cp -r /tmp/okou-skills/extract-template/presentation .claude/skills/presentation-extract-template
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Presentation Extract Template 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 Presentation Extract Template 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 Presentation Extract Template 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.

Extract a presentation template

Preserve the original source and save its style, layouts and assets as a reusable HTML package. Extraction uses package checks; deck QA and targeted visual review are for later generation. Rebuild samples only when requested, within that scope. Write package instructions in English; preserve source content and the requested presentation language.

Files in this guideStageIncluded in the saved template
This SKILL.md, references/extract/Extract and record the sourceNo
references/generate/Generate, check and deliver new decksYes, at the same paths
Source helpers in scripts/ and tools/verify-package.mjsPrepare source files, assemble and check the packageNo

The installer copies the generation guides and runtime by an explicit file list. The package's own SKILL.md is its generation entrypoint.

1. Inspect the source

InputExtraction route
HTML file, directory or URLInventory the HTML and reuse its fragments, CSS and assets.
PDF deckPrepare, inspect and assemble through the HTML extraction contract; after PACKAGE_VALID, continue at step 5.
PPTX / PPTRead available structure/style metadata together with the rendered pages.
Slide imagesInfer layouts and styling from the supplied pages.

Keep the original as evidence. Source parsing/conversion and the independent page capture can run together.

Use a new <work-dir> beside the template package. All commands below use fixed subdirectories: source-pages/ for original pages, previews/ for contact sheets and review images, and upload/ for the upload PDF.

Reuse existing ordered captures or supplied slide images at <work-dir>/source-pages/page-001.png, etc. Otherwise, render once:

bash
okou presentation screenshot --input <source.ppt|source.pptx|source.pdf|source.html|html-dir|url> --out <work-dir>/source-pages

Review every page through contact sheets; open individual captures only for details that are unclear at that size. The helper installs Pillow if needed and writes previews to their own directory. For HTML/images without an original PDF, create the upload PDF in the same pass:

bash
python3 scripts/source-derivatives.py --work-dir <work-dir> --contact-sheet --pdf

For PPT/PPTX/PDF, omit --pdf: the original already supplies the upload source. For HTML, use --slides "<slide-selector>" when the source uses a different slide selector from .deck > .slide; use --slides none for one page per HTML file. Read computed CSS once per distinct style, including inherited styles and SVG. Record the original entry file or URL in design-system.md.

Record the filename, aspect ratio, page count, page roles and recurring structures. Check typography against source metadata and label visual estimates as inferred. Archive original-source captures, never reconstruction images.

2. Record the design system

In design-system.md, record typography, foreground/background colours, spacing, safe areas, borders, image crops, required chrome and reusable backgrounds/decorations. Distinguish source observations from inferred adaptations, including unit conversion and aspect-ratio mapping to 16:9.

Use typography.md for source font mapping and layout-inventory.md for source coverage and reusable background recipes.

3. Build the editable package

text
<template-slug>/
  SKILL.md
  design-system.md
  layouts/README.md
  layouts/source-index.json
  layouts/source/<name>.html
  layouts/_shell.html
  styles/template.css
  assets/
  library/                  # supplied neutral layouts and Markdown catalogue
  scripts/                  # supplied fitter and navigation
  tools/                    # source configuration and generation QA
  references/generate/      # generation-only guidance

Create the source index and fragments in one batch; preserve every distinct composition and background variant. Equivalent structures may share a layout with explicit source-page coverage. Keep text, shapes, tables and ordinary charts editable in HTML/CSS/SVG. Use package-relative assets and .deck > .slide > .stage, with a .fit root per slide. Extra data-fit-region hooks are only for genuinely constrained subareas; ordinary flow text needs no extra measurement boundary. All content shares one scale.

layouts/_shell.html is the shared HTML document with one .deck insertion point. The installer supplies a pure-black viewer surround with centered, proportional canvas scaling. Keep source dimensions, backgrounds and styling on .stage and inside it; no separate viewport or resize script is needed. Its .deck can be empty or retain source examples; no need to assemble every source page in it. Complete source-page coverage belongs in source-index.json. Source fragments use paths relative to layouts/, where they are inserted into the shell.

Write tools/qa-config.json with source-derived safe areas, required chrome and approved font fallbacks; follow qa-config.md. Use the runtime hooks when creating the source HTML, so later generation can reuse it directly. Then run from this guide's directory:

bash
node scripts/install-qa.mjs --package <template-slug>

The installer supplies the library, tools, references and shared runtime, including the viewport frame and four-arrow navigation. Keep neutral library layouts separate from the source index; adapt only selected layouts during later generation.

Write the package's SKILL.md with its template identity, source/language caveats, links to design-system.md and layouts/source-index.json, and a Markdown link to references/generate/workflow.md. That supplied workflow contains the generation steps and routes to the other guides; no need to duplicate them. Keep layouts/README.md a short path map linking back to the package's SKILL.md. When refreshing an older package, update its entrypoint links to references/generate/. After READY_TO_PUBLISH, direct future runs to the platform publishing and delivery instructions. Keep QA restrictions scoped to content QA.

4. Check the package

From this guide's directory, run the package verifier. It checks slide/stage/fit structure, generation links, source coverage/configuration, library resources and capture inventory, and rejects extraction files mixed into the saved package:

bash
node tools/verify-package.mjs \
  --package <template-slug> --source-pages <work-dir>/source-pages

No generation QA, interaction pass or runtime test-suite run is needed during extraction. Only for a specific reconstruction risk, compare one or two unchanged compositions with their source captures, using --out <work-dir>/previews/reconstruction for any check screenshots.

5. Save the Custom Template

After the package check passes, prepare the upload source.

Use the original .ppt, .pptx or .pdf as --source. For HTML/images, reuse <work-dir>/upload/source.pdf from step 1. Record this conversion in design-system.md; the package retains editable layouts and assets. No reconstructed export is needed.

bash
okou user-template publish --kind presentation --title "<template name>" \
  --source <original.ppt|original.pptx|original.pdf|work-dir/upload/source.pdf> \
  --pages <work-dir>/source-pages \
  --package <template-slug>

Confirm successful saving and retain any returned template reference; if publishing fails, report that the template was not saved.

Bundled files

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

and 25 more files.

Frequently asked questions

What does the Presentation Extract Template AI skill do?

Extract source style and editable HTML layouts from HTML decks (files, directories or URLs), PPTX, PPT, PDF or slide images, then save a reusable Custom Template.

Why use Presentation Extract Template on TypingMind?

Because you install it once and use it with any model. Presentation Extract Template 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 Presentation Extract Template in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/okou-ai/okou-skills/tree/main/extract-template/presentation. 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 Presentation Extract Template?

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 Presentation Extract Template?

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

Is the Presentation Extract Template AI skill free?

It is published on GitHub by okou-ai. Check the repository for licensing terms. 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 👇