Agent skill

Slide Design

by humantonylee in humantonylee/future-slide

Shared slide-style extraction skill for GPT-image and HTML slide workflows.

Apache-2.0Auto-check passedDocuments & Office

Install Slide Design

skills CLI
$ npx skills add humantonylee/future-slide --skill slide-design -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install humantonylee/future-slide slide-design --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/humantonylee/future-slide.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/slide-design .claude/skills/slide-design && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
slide-design
GitHub stars
150
Token cost
~2.6k tokens
SKILL.md length
1,070 words
Files
1
Skills in repo
10
Repo updated
First seen
Licence
Apache-2.0

At a glance

Shared slide-style extraction skill for GPT-image and HTML slide workflows.

  • Works in 7 steps: Composition → Typography → Color and contrast → …
  • Tasks that involve Slides and decks
  • SKILL.md covers Trigger commands, Recommended inputs, Primary goal and What DESIGN.md means in this…, plus 9 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Slide Design is an agent skill from humantonylee/future-slide. Shared slide-style extraction skill for GPT-image and HTML slide workflows. Converts one or more reference visuals into a reusable DESIGN.md focused on presentation design, layout systems, and implementation-ready constraints.

Its SKILL.md is about 2.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Documents & Office, covering Slides and decks and Design tokens. The repository describes itself as: The workflow for disciplined slide generation. Extract design, plan the deck, render with consistency. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Slides and decks
  • Tasks that involve Design tokens

Example prompts

  • “/slide-design”

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. Composition
  2. Typography
  3. Color and contrast
  4. Layout system
  5. Page flow and zone architecture
  6. Components
  7. Data-viz language

What it can do on your machine

Read from SKILL.md and the folder at commit 26914e8. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Slide Design loads about 2.6k tokens when it runs. Until then it costs about 60 tokens; SKILL.md has 1,070 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~60
When it runs · the whole SKILL.md, loaded when a task matches
~2.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from humantonylee/future-slide at commit 26914e8, republished under its Apache-2.0 licence (© humantonylee). 1,070 words, ~2,622 tokens.

Download SKILL.mdSave it as .claude/skills/slide-design/SKILL.md (or your agent's skills folder).
name
slide-design
description
Shared slide-style extraction skill for GPT-image and HTML slide workflows. Converts one or more reference visuals into a reusable DESIGN.md focused on presentation design, layout systems, and implementation-ready constraints.

SLIDE DESIGN — EXTRACT DESIGN.md FROM REFERENCE VISUALS

You are an elite presentation design analyst. Your job is to convert a reference slide image into a reusable DESIGN.md that captures the visual system of the slide.

This is not an OCR task. This is not a slide-summary task. This is not a content-rewrite task.

Your job is to infer the design language of the reference slide so later steps can generate new slides in the same family.

Trigger commands

Use this skill for shared design extraction in either output mode:

  • $slide-design
  • "extract DESIGN.md from this reference slide"
  • "create a slide design system"
  • "analyze this reference for GPT image slides"
  • "analyze this reference for HTML slides"

Use downstream workflow skills for later stages:

  • GPT image: $gpt-image-slide-plan, $gpt-image-slide-prompt, $gpt-image-slide-render
  • HTML: $html-slide-plan, $html-slide-prompt, $html-slide-render

Use the strongest available design reference, not only a single screenshot. Recommended inputs include:

  • reference slide image(s)
  • exported slide images from a reference deck
  • PDF files that contain key colors, structured tables, charts, report headers, source notes, disclosure blocks, or other report-like layout systems

When a PDF is available, prefer report PDFs or structured analytical documents over loose text documents. A useful PDF should expose the deck/report color system and repeated information architecture: table density, chart treatment, section headers, footers, labels, and evidence layout. Treat those visual and structural signals as design evidence, while still avoiding content summary.

Primary goal

Extract a DESIGN.md that captures:

  • visual theme and atmosphere
  • palette and contrast behavior
  • typography hierarchy
  • spacing and density
  • layout grammar
  • layout placement by zone
  • header / body / footer structure
  • title page / body page / end page flow
  • chart/table treatment
  • infographic behavior including card systems, icon-led cards, and diagram flow
  • graphic and image treatment
  • implementation notes for GPT-image and HTML/CSS output modes when relevant
  • cross-slide consistency rules
  • anti-patterns to avoid

What DESIGN.md means in this workflow

DESIGN.md is a semantic presentation design specification. It is the slide deck equivalent of a design system or style guide. It should be reusable for new content, not bound to the specific wording of the original slide.

The output must describe how the slide is designed, not what the slide says.

Output mode notes

If the next stage is GPT image generation, emphasize prompt-safe visual constraints: composition, hierarchy, page families, chart/table treatment, icon rules, and anti-patterns that image generation must avoid.

If the next stage is HTML slides, also include browser-implementable guidance: approximate color tokens, type scale, aspect-ratio behavior, CSS layout constraints, overflow risks, and responsive or print/PDF considerations.

Observation hierarchy

When analyzing the reference slide image, inspect in this order:

1. Composition
  • title block position
  • content balance
  • left/right or top/bottom dominance
  • large whitespace zones
  • edge anchoring
  • asymmetry vs symmetry
2. Typography
  • title size and weight
  • subtitle and body treatment
  • casing conventions
  • alignment logic
  • list styling
  • line length
  • numeric emphasis
3. Color and contrast
  • background color
  • text colors
  • accent colors
  • separator and border usage
  • chart colors
  • highlight colors
  • whether the design is muted, corporate, editorial, premium, technical, playful, dense, restrained
4. Layout system
  • likely grid
  • margin discipline
  • alignment baselines
  • repeated structural zones
  • likely master-slide constraints
5. Page flow and zone architecture
  • title page structure
  • body page structure
  • end page / CTA structure
  • header / body / footer rhythm
  • what content typically lives in each zone
  • how transitions between page families are handled visually
6. Components
  • headers
  • callout boxes
  • tables
  • charts
  • icons
  • image masks
  • labels
  • annotations
  • footnotes
  • source text
7. Data-viz language
  • bar vs line vs scatter vs waterfall tendencies
  • axis treatment
  • gridline visibility
  • legends
  • annotation blocks
  • chart title/subtitle structure
  • infographic card patterns
  • icon + label + metric treatment
  • process / timeline / diagram flow behavior

Mandatory distinction: observed vs inferred

You must clearly separate:

  • Observed — directly visible in the image
  • Inferred — likely system behavior that is not fully visible but can be reasonably generalized

Never state an inferred rule as if it were directly visible.

Show full SKILL.md (437 more words)Show less

Strict anti-hallucination rules

  • Do not invent a brand name, font name, or exact color hex unless you can justify it from the image.
  • If exact font identification is uncertain, describe the class: “neo-grotesk sans”, “humanist sans”, “editorial serif”, “condensed display sans”.
  • If exact hex is uncertain, provide an approximate hex and mark it as approximate.
  • Do not copy the textual content of the slide except when it is useful to explain structure.
  • Do not infer deck narrative from one image. Infer slide design only.

Tightening rules

You must aggressively resist generic slide-design defaults.

  • Do not describe layouts vaguely. Name exact placement logic by zone and relative dominance.
  • Do not accept “one title + random boxes” as a design system unless the reference truly shows that.
  • Do not ignore table-heavy or report-heavy slide families just because they are less visually exciting.
  • Detect whether the deck is presentation-led, report-led, analyst-note-led, or hybrid, and encode that explicitly.
  • Detect whether headers and footers are branded/systematic rather than decorative, and encode that as a rule.
  • If the reference uses tables, report headers, stock-note footers, disclosure areas, or chart captions repeatedly, treat them as part of the visual system, not as throwaway content.
  • Explicitly note if titles are kept wide and short versus narrow and over-wrapped.
  • Explicitly note if infographic elements are tightly disciplined and repetitive rather than expressive or varied.
  • Explicitly call out cheap defaults to avoid: narrow 5-6 line headlines, arbitrary badge spam, empty grid gaps, floating icons with no semantic role, decorative diagrams with no analytical function.

Required output format

Return exactly one DESIGN.md using this structure:

markdown
# Design System: [Concise style name]

## 1. Design Intent
- **Observed from reference:**
- **Inferred but not directly visible:**
- **Overall impression:**
- **Appropriate use cases:**

## 2. Color System
- **Canvas / background:**
- **Primary text:**
- **Secondary text:**
- **Accent 1:**
- **Accent 2 (only if clearly present):**
- **Dividers / borders:**
- **Chart colors:**
- **Banned / avoid:**

## 3. Typography System
- **Title style:**
- **Section header style:**
- **Body style:**
- **Caption / source / footnote style:**
- **Numeric emphasis style:**
- **Observed casing rules:**
- **Observed line-length behavior:**

## 4. Layout Families
- **Cover / opener:**
- **Section divider:**
- **Insight / claim slide:**
- **Chart / data slide:**
- **Comparison slide:**
- **Process / timeline slide:**
- **Closing / CTA slide:**

## 5. Flow Architecture
- **Title page flow:**
- **Body page flow:**
- **End page flow:**
- **Header / body / footer structure:**
- **Header zone placement rules:**
- **Body zone placement rules:**
- **Footer zone placement rules:**

## 6. Grid, Alignment, and Spacing
- **Outer margins:**
- **Column behavior:**
- **Text alignment:**
- **Whitespace philosophy:**
- **Density level:**
- **Object anchoring rules:**

## 7. Components
- **Title block:**
- **Subtitle / kicker:**
- **Bullets / key points:**
- **Cards / callouts:**
- **Tables:**
- **Charts:**
- **Legends / labels:**
- **Icons / illustrations / photography:**
- **Icon placement and usage rules:**
- **Infographic cards / metric cards:**
- **Diagram / flow modules:**

## 8. Data Visualization Language
- **Preferred chart families:**
- **Axis / gridline treatment:**
- **Labeling style:**
- **Annotation style:**
- **When to avoid charts:**
- **Infographic composition style:**
- **Icon-led data communication:**
- **Diagram flow direction and connector behavior:**

## 9. Imagery and Graphic Treatment
- **Image crop / masking:**
- **Use of gradients / fills:**
- **Use of shapes / panels / bands:**
- **Use of texture / shadows:**

## 10. Slide-System Rules
- **What repeats across most slides:**
- **What should vary cautiously:**
- **Body-slide layout discipline:**
- **What must remain consistent across the deck:**
- **What stays consistent across title/body/end pages:**
- **How icons/infographics should repeat across the deck:**

## 11. Anti-Patterns
- [list]

Body-slide discipline rule

You must explicitly define what “body slides” should look like across the deck. This is critical. The next skills will reuse these layout families for most interior slides.

Placement and infographic rule

You must explicitly describe:

  • where major elements are placed on the page, not just what components exist
  • whether icons are decorative, explanatory, metric-supporting, or diagram-driving
  • whether infographics are card-based, chart-led, icon-led, or flow-diagram-based
  • how header, body, and footer zones are reused across title, body, and end pages
  • how dense analytical tables, charts, and disclosure/report modules fit into the page family system when they appear

Mandatory completion check

Before finalizing DESIGN.md, verify:

  1. You described design, not content.
  2. You separated observed vs inferred.
  3. You included reusable layout families.
  4. You included chart style and table style if visible or likely relevant.
  5. You stated body-slide consistency rules.
  6. You listed anti-patterns.
  7. You defined header / body / footer flow.
  8. You covered title page, body page, and end page behavior.
  9. You described icon usage and infographic / diagram behavior.

© humantonylee, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/slide-design of humantonylee/future-slide.

Open the folder on GitHubat commit 26914e8

Compare with similar skills

Slide Design next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Slide Design compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Slide Design this skillhumantonylee/future-slide150—~2.6kAutomated safety check: PassApache-2.0
Brand and Design Toolkitnextlevelbuilder/ui-ux-pro-max-skill134k1 repos~3.5kAutomated safety check: PassMIT
Design SystemOhh-889/skyroc79511 repos~1.7kAutomated safety check: PassMIT
Image To Editable Pptningzimu/image-to-editable-ppt-skill2.8k—~4.3kAutomated safety check: PassMIT
Slidesfcakyon/claude-codex-settings1.2k1 repos~1.1kAutomated safety check: PassMIT
Ppt Image FirstNyxTides/ppt-image-first1.2k—~1.6kAutomated safety check: PassApache-2.0

Similar skills

  • Brand and Design Toolkit

    nextlevelbuilder/ui-ux-pro-max-skill

    Bundles design tasks behind one skill: brand identity, tokens, UI styling, logos, corporate identity mockups, slides, banners, icons and social images.

    134k GitHub starsUsed in 1 repo~3.5k tokens
    Media & CreativeAuto-check passed
  • Design System

    Ohh-889/skyroc

    Token architecture, component specifications, and slide generation.

    795 GitHub starsUsed in 11 repos~1.7k tokens
    Frontend & DesignAuto-check passed
  • Image To Editable Ppt

    ningzimu/image-to-editable-ppt-skill

    Rebuild slide images, scanned or image-based PPT/PPTX files, and PDF decks into object-level editable PowerPoint (.pptx), preserving speaker notes when supplied.

    2.8k GitHub stars~4.3k tokensUpdated 23 days ago
    Documents & OfficeAuto-check passed
  • Slides

    fcakyon/claude-codex-settings

    Create and edit presentation slide decks (.pptx) with PptxGenJS, bundled layout helpers, and render/validation utilities.

    1.2k GitHub starsUsed in 1 repo~1.1k tokens
    Documents & OfficeAuto-check passed
  • Ppt Image First

    NyxTides/ppt-image-first

    Build presentation plans for PPT / slides / decks through a conversation-first workflow, then propose multiple visual directions with preview images before writing deck specs.

    1.2k GitHub stars~1.6k tokensUpdated 5 mo ago
    Documents & OfficeAuto-check passed
  • Vibe to Agentic Engineering Framework

    shanraisshan/claude-code-best-practice

    Explains the conceptual model behind a presentation on moving from unstructured vibe coding to fully configured agentic engineering, including its 4-level scoring system and slide conventions.

    67k GitHub stars~3.3k tokensUpdated today
    Documents & OfficeAuto-check passed

More from humantonylee/future-slide

All 10 skills in this repo
  • Tightened Slide

    humantonylee/future-slide

    Build single-file HTML horizontal-swipe presentation decks using the Tightened Slide process with locked layouts, strict grid rules, theme presets, image slot discipline, and validation.

    150 GitHub stars~1.3k tokensUpdated 4 mo ago
    Auto-check passed
  • HTML Slide Render

    humantonylee/future-slide

    Final HTML slide stage. An agent skill from humantonylee/future-slide.

    150 GitHub stars~377 tokensUpdated 4 mo ago
    Auto-check passed
  • Gpt Image Slide Plan

    humantonylee/future-slide

    Persuasive GPT image deck-planning skill. An agent skill from humantonylee/future-slide.

    150 GitHub stars~2.6k tokensUpdated 4 mo ago
    Auto-check passed
  • Gpt Image Slide Prompt

    humantonylee/future-slide

    Page-level GPT image slide prompting skill. An agent skill from humantonylee/future-slide.

    150 GitHub stars~2k tokensUpdated 4 mo ago
    Auto-check passed
  • Gpt Image Slide Render

    humantonylee/future-slide

    Render slide images sequentially from DESIGN.md and slide prompt JSON using Codex native image generation, saving each output into the workspace with page-number filenames.

    150 GitHub stars~966 tokensUpdated 4 mo ago
    Auto-check passed
  • Gpt Image Slide

    humantonylee/future-slide

    End-to-end GPT image slide workflow. An agent skill from humantonylee/future-slide.

    150 GitHub stars~397 tokensUpdated 4 mo ago
    Auto-check passed

Questions about Slide Design

What does Slide Design do?

Shared slide-style extraction skill for GPT-image and HTML slide workflows. Slide Design is an agent skill from humantonylee/future-slide. Shared slide-style extraction skill for GPT-image and HTML slide workflows.

When should I use Slide Design?

Slide Design fits situations like: tasks that involve Slides and decks; tasks that involve Design tokens.

How do I install Slide Design in Claude Code?

Run `npx skills add humantonylee/future-slide --skill slide-design -a claude-code`. Or copy the skill folder (skills/slide-design in humantonylee/future-slide) into .claude/skills/slide-design in your project. Claude Code loads it when a task matches its description.

How do I install Slide Design in Codex?

Run `npx skills add humantonylee/future-slide --skill slide-design -a codex`. Or copy the skill folder (skills/slide-design in humantonylee/future-slide) into .agents/skills/slide-design in your project. Codex loads it when a task matches its description.

Can I use Slide Design in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add humantonylee/future-slide --skill slide-design -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/slide-design, .gemini/skills/slide-design, .github/skills/slide-design and .opencode/skills/slide-design in your project.

What does Slide Design need to run?

SKILL.md names no scripts, command-line tools or credentials: Slide Design is instructions for the agent only.

Does Slide Design access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Slide Design safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Slide Design use?

Slide Design is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Slide Design use?

About 2.6k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Slide Design?

Skills that share tags, products or a category with Slide Design: Brand and Design Toolkit (nextlevelbuilder/ui-ux-pro-max-skill, 134k stars), Design System (Ohh-889/skyroc, 795 stars), Image To Editable Ppt (ningzimu/image-to-editable-ppt-skill, 2.8k stars) and Slides (fcakyon/claude-codex-settings, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Slide Design?

humantonylee (a GitHub user) maintains it in humantonylee/future-slide, which has 150 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on June 2, 2026.

Source: humantonylee/future-slide on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.