Deck Keynote

Organization

FluidForm-ai

deck-keynote

Build narrative-driven, speaker-centric keynote decks (20-35 slides, default 28) for conference talks, TED-style presentations, and standalone speaker decks. Triggers, "keynote", "keynote deck", "conference talk", "TED-style talk", "speaker deck". Inherits full pipeline from deck-builder core; type-specific spine, visual patterns, and demo recipes defined here. Keynotes are the most elastic and visual of all deck types, every slide requires a visual anchor, one idea per slide, mandatory opening hook.

Overview

Publisher	FluidForm-ai
Repository	`fluiddocs-deck-builder`
Skill name	`deck-keynote`
Stars	33
Forks	2
Bundled files	3
License	MIT
Links	Skill source Repository Homepage

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.
3 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by FluidForm-ai on GitHub. Read the source before you install it.

Installation

Install the Deck Keynote AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

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.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/FluidForm-ai/fluiddocs-deck-builder/tree/main/skills/deck-keynote
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.

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/FluidForm-ai/fluiddocs-deck-builder.git /tmp/fluiddocs-deck-builder
mkdir -p .claude/skills
cp -r /tmp/fluiddocs-deck-builder/skills/deck-keynote .claude/skills/deck-keynote

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Deck Keynote 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 Deck Keynote 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 Deck Keynote 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.

Deck Keynote, a type pack for the `deck-builder` core

Keep your process invisible. Do not narrate the skill architecture (the "type pack" and "core" split) or announce which files you are reading. Skip preambles like "I'll start by reading...". Gather what you need, then build the deck.

This skill builds conference keynotes, TED-style talks, and speaker-centric presentations. It is a thin type pack that inherits the full pipeline from the deck-builder core skill. Read deck-builder/SKILL.md first for the 5-phase pipeline, 3-reviewer gate, fixed-canvas shell, brand-tokens methodology, and learnings log. This file only declares what's keynote-specific.

What's type-specific (defined in this pack)

References in references/:

content-spine.md, the canonical 28-slide keynote spine (flexible 20-35 range)
visual-components.md, per-slide visual pattern catalog, emphasizing imagery and large-type design
demo-patterns.md, optional interactive demo recipes suited to live talks (kept as a visual reference; most keynotes ship without a demo slot)

Declarations the brief locks in for keynote builds:

deck_type: keynote
Target slide count: 28 (flexible 20-35)
Catalog directory ($CATALOG): user-specified output directory
Forbidden class-name leaks: (empty)
Demo: optional (off by default)
Density mode: live-presentation by default; ask the user whether the deck is for a live talk (sparse, image-led, 45s/slide pacing) or for async reading (denser text, self-explanatory slides)

Phase 1 intake (keynote-specific batched questions)

When the trigger fires, the agent runs Phase 1 from deck-builder/SKILL.md. In addition to the generic intake, ask these keynote-specific questions in a single batched message:

Talk length: how many minutes is the speaking slot? (Drives slide count: ~45 seconds per slide.)
Density mode: is this deck for a live presentation (sparse, image-led, speaker carries the words) or for async reading (denser text, the slides explain themselves without a speaker)? This drives type sizes, text-per-slide, and whether speaker notes are inline or hidden.
Thesis in one sentence: the single idea every slide must support.
Demo: does the talk include a live interactive demo moment? (Default: no.)
Speaker brand tokens: any existing speaker site, talk page, or brand kit to mirror?

Batch these into one message. Do not interrogate slide by slide.

Phase 2 auto-preview (run on turn 1 of Phase 2)

Before writing the full deck, the agent must auto-generate 3 first-slide previews and let the user pick a direction. Do not ask the user how to choose a style; pick the 3 from deck-builder/references/style-presets.md:

Safe preset: a clean, conference-default cover (full-bleed photo + large title, dark text on light bg or vice versa).
Bold preset: a high-contrast bold-type cover (single thesis phrase at 240pt+, solid color background).
Wildcard preset: an unexpected aesthetic (asymmetric layout, oversized monogram, or an artistic visual that matches the speaker's brand if known).

Steps:

Write the 3 first-slide HTML previews to .skill-temp/keynote-preview-{safe,bold,wildcard}.html.
Open all 3 in the user's browser.
Ask, "Which style do you want? A, B, C, or mix elements from these?"
Only after the user picks does the full Phase 2 build begin.

Autonomous run exception: for autonomous runs without a user browser to open previews in, skip the visual preview step. Instead, document the preset choice (palette, typography, character) as a code block in the brief and proceed directly to full build. The user can iterate on the preset post-build via the inline-edit module if needed.

What this type pack inherits from `deck-builder` core

Pipeline: deck-builder/SKILL.md, 5-phase Plan, Build, Review, Release, Learn; 3 reviewer gates.

Phase 1 artifact: deck-builder/references/build-brief-template.md, fill in deck_type: keynote plus the other keynote-specific fields listed above.

Phase 2 implementation:

deck-builder/references/shell-pattern.md, chrome-free nav shell + 1440×810 canvas scale + inline-edit module
deck-builder/references/style-presets.md, the preset catalog used by the auto-preview step above
deck-builder/references/icon-library.md, inline SVG icon set replacing every emoji
deck-builder/references/brand-methodology.md, source-verify brand tokens protocol + logo safety + nested-subpath rule

Phase 3 reviewer specs (all in deck-builder/reviewers/):

brand.md, copy.md, layout.md

Phase 5 learnings: deck-builder/references/learnings-log.md (shared; filter by Deck type = keynote).

Output expectations (keynote-specific)

Phase 4 deliverable:

The .html file at <CATALOG_DIR>/<Speaker-Name>-Keynote.html or user-specified path.
The brief alongside it: <Speaker-Name>-Brief.md.
The absolute path to the file (or a tool-specific link such as computer://... when the agent runs in an environment that resolves it).
A one-paragraph design/content summary.
The compact audit report (3 reviewers, one line each).

Do not write walls of explanation. The file is the deliverable.

Hard rules specific to keynote decks

No slide is all text. Every slide has a visual anchor, a full-bleed image, a chart, a single word at 120pt+, or a bold design element. Slides support the speaker; they do not substitute for the speaker.
One idea per slide. Each slide carries one main thought, one story beat, or one data point. No slide is a paragraph; no slide lists five unrelated ideas.
Opening hook is mandatory. Slide 2 (immediately after cover) must hook the audience with a story, a striking stat, or a provocative statement. Never start with "Hi, I'm X" or an agenda slide.
45 seconds per slide. Keynote reads at a strict 45 seconds per slide. A 28-slide deck is a ~21-minute talk; a 20-slide deck is ~15 minutes; a 35-slide deck is ~26 minutes. Content and slide count must align with the intended talk length.
Density mode discipline. If live-presentation mode, keep text minimal; if async-reading mode, slides must stand alone without a speaker. Do not mix modes within a single deck.
No emoji on any slide. All decorative elements use the icon library (deck-builder/references/icon-library.md); emoji are forbidden.

Maintained by FluidDocs. Source: https://github.com/FluidForm-ai/fluiddocs-deck-builder. MIT licensed.

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 Deck Keynote AI skill do?

Why use Deck Keynote on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/FluidForm-ai/fluiddocs-deck-builder/tree/main/skills/deck-keynote. 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 Deck Keynote?

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 Deck Keynote?

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

Is the Deck Keynote AI skill free?

Yes. It is published on GitHub by FluidForm-ai 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

Grill Me

mattpocock

CommunityPopular

A relentless interview to sharpen a plan or design.

Brand Discovery

Use when a brand needs to discover or articulate its identity through structured multi-session interviews. Covers purpose, positioning, audience, personality, voice, narrative, and founder-brand…

Competitive Platform Analysis

affaan-m

CommunityPopular

Use when scoping a competitive landscape — identifying, categorising, and score-filtering a competitor set before any benchmarking begins. Decides who counts as a competitor, which tier they belong…

Frontend Slides

Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a…

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Deck Keynote, a type pack for the deck-builder core

What's type-specific (defined in this pack)

Phase 1 intake (keynote-specific batched questions)

Phase 2 auto-preview (run on turn 1 of Phase 2)

What this type pack inherits from deck-builder core

Output expectations (keynote-specific)

Hard rules specific to keynote decks

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now

Deck Keynote, a type pack for the `deck-builder` core

What this type pack inherits from `deck-builder` core