Lecture Deck logo

Lecture Deck

Community
RinDig
lecture-deck

Build a high-quality animated HTML lecture or workshop deck from a storyboard, in any theme. Use whenever someone asks for a slide deck, session deck, workshop slides, lecture screens, a run-of-show deck, or speaker notes for a talk, especially one delivered live (in a room or on Zoom) with breakouts, share-screen moments and handoffs. Produces one self-contained HTML file with keyboard navigation, speaker notes, overview, print, copy-to-clipboard prompts and a mode strip on every screen, plus PNG renders for checking. Ships with three themes and a kit for diagrams, animation, a recurring cast of figures, and a register checker. Not for pitch decks that must be PowerPoint, or for data dashboards.

Overview

PublisherRinDig
Repositorylecture-deck-skill
Skill namelecture-deck
Stars
52
Forks
10
Bundled files
21
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.

  • 21 bundled files

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

  • Open source

    Published by RinDig on GitHub. Read the source before you install it.

Installation

Install the Lecture Deck 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/RinDig/lecture-deck-skill/tree/master
  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/RinDig/lecture-deck-skill.git \
  .claude/skills/lecture-deck
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Lecture Deck 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 Lecture Deck 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 Lecture Deck 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.

Lecture deck

One claim per screen, one picture that shows the mechanism, one caption, and everything else in the notes. The deck is built from Python screen modules against a theme, rendered to PNG for checking, and checked for register before it leaves.

When you get a request

  1. Ask for or infer the room: who, how many, where, how long, what they already know, who speaks and who runs the room. If a run of show exists (a host's timings), take it as given.
  2. Write storyboard.md first, using reference/storyboard-template.md. Goal, argument as numbered claims, run of show with modes, then one entry per screen with its picture and its bridge. Read the headers alone as an argument; then the bridges alone. Fix order here.
  3. For anything over ten screens, critique the storyboard before building (flow, audience level, whether each picture explains its claim, coverage of the brief). Fold the changes in.
  4. Scaffold and build the screens (reference/workflow.md, reference/kit-api.md). For a long deck split screens into modules and build them in parallel.
  5. Render every screen and look at every contact sheet. Fix overflow, clipping, unreadable text, pictures that decorate instead of explain.
  6. Run the register check with --dump and read the text flat. Fix, re-render what changed, repeat until clean.
  7. Deliver the HTML, name what is placeholder (dates, names, a platform that may not be ready), and say how to present it (arrows, N, O, B, F, Print).

The rules that make it good

  • The load rule. Header is the claim. One picture. At most one caption. Footnotes only for verified quotations. Cues only on exercise and logistics screens. Depth goes into SAY and IF THE ROOM IS AHEAD, word for word.
  • Pictures show mechanisms. A picture that could sit under any header is decoration. Draw the actual thing: the chat, the file, the folder, the person bending, the bins filling. Build it up in the order the speaker will talk, with anim, draw, grow, rise, slide_in, drop. Complex is fine when it explains.
  • One cast. Use figure, machine, rigid for every person, system and rigid box so the audience meets the same drawings all deck.
  • Roles, not colours. HIGHLIGHT is the subject and now; COOL the system; WARM what is written; SOFT the other party; NEUTRAL gaps; LIVE only the mode dot. Any theme then works.
  • The mode strip. Every screen starts with mode(kind, label, who): slides, share, breakout, solo, handoff, break. The audience always knows who has the screen; the speaker always knows when to swap apps or hand over. The run-of-show bar uses the same colours.
  • Register. Open on the claim, full sentences, explain rather than perform. No reveal lines, no "not X but Y" beyond one per deck, no fragments that land a point, no forced triads, no dashes, no stock AI words, no method vocabulary on audience screens. Verified quotations only. Examples one per audience segment, same order every time. Details in reference/writing.md.
  • Notes are the lecture. SAY in the speaker's voice, DO NOW with times for the host, LOOK FOR, IF STUCK, IF THE ROOM IS AHEAD, and a BRIDGE that leads into the next header.
  • Check at real size. The renderer's static mode shows every animated element at its final state. Faults invisible in a browser pane show at 1280×720.

Commands

python scripts/new_deck.py path/to/deck --title "Session one" --theme eduba --left "PROGRAMME" --right "SESSION 1"
python scripts/build_deck.py path/to/deck            # writes deck/out/<name>.html
python scripts/build_deck.py path/to/deck --theme slate   # a second file in another theme
python scripts/render.py path/to/deck/out/<name>.html     # PNG per screen + contact sheets, static mode
python scripts/check_text.py path/to/deck/out/<name>.html --dump

Requirements: Python 3.9+, a Chromium for rendering (Playwright's, Chrome or Edge; or set DECK_CHROME), Pillow for contact sheets. Fonts load from Google Fonts at open time; the theme carries a fallback stack.

Files

  • scripts/deckkit/kit.py the helpers, constants, cast, common pictures. scripts/deckkit/shell.py the page shell. scripts/deckkit/themes/*.json the themes.
  • reference/workflow.md the loop · reference/writing.md the register · reference/kit-api.md every helper · reference/themes.md roles and how to make a theme · reference/storyboard-template.md.
  • examples/demo/ a nine-screen deck that exercises most of the kit; build it in each theme to see the range.

Bundled files

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

Frequently asked questions

What does the Lecture Deck AI skill do?

Build a high-quality animated HTML lecture or workshop deck from a storyboard, in any theme. Use whenever someone asks for a slide deck, session deck, workshop slides, lecture screens, a run-of-show deck, or speaker notes for a talk, especially one delivered live (in a room or on Zoom) with breakouts, share-screen moments and handoffs. Produces one self-contained HTML file with keyboard navigation, speaker notes, overview, print, copy-to-clipboard prompts and a mode strip on every screen, plus PNG renders for checking. Ships with three themes and a kit for diagrams, animation, a recurring c...

Why use Lecture Deck on TypingMind?

Because you install it once and use it with any model. Lecture Deck 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 Lecture Deck in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/RinDig/lecture-deck-skill/tree/master. 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 Lecture Deck?

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 Lecture Deck?

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

Is the Lecture Deck AI skill free?

Yes. It is published on GitHub by RinDig 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 👇