Frontend Slides

Create zero-dependency, animation-rich HTML presentations that run entirely in the browser. Single self-contained HTML files — no npm, no build tools, no frameworks. Works offline, renders in 10 years.

Style presets: Bold Signal, Electric Studio, Creative Voltage, Dark Botanical, Notebook Tabs, Pastel Geometry, Split Pastel, Vintage Editorial, Neon Cyber, Terminal Green, Swiss Modern, Paper & Ink

For full style details and CSS specs: read references/STYLE_PRESETS.md when needed.

Phase 0: Detect Mode

Determine the user's intent:

• Mode A — New Presentation: Create slides from scratch → go to Phase 1
• Mode B — PPT Conversion: Convert an existing .ppt/.pptx file → go to Phase 4
• Mode C — Enhance Existing: Improve an existing HTML presentation → read the file, then enhance (always maintain viewport fitting)

CRITICAL: Viewport Fitting (Non-Negotiable)

Every slide MUST fit exactly within the viewport. No scrolling within slides, ever.

Content Density Limits Per Slide

Content exceeds limits → Split into multiple slides.

Required CSS for Every Presentation

html, body { height: 100%; overflow-x: hidden; }
html { scroll-snap-type: y mandatory; scroll-behavior: smooth; }

.slide {
    width: 100vw;
    height: 100vh;
    height: 100dvh;
    overflow: hidden;
    scroll-snap-align: start;
    display: flex;
    flex-direction: column;
    position: relative;
}

.slide-content {
    flex: 1;
    display: flex;
    flex-direction: column;
    justify-content: center;
    max-height: 100%;
    overflow: hidden;
    padding: var(--slide-padding);
}

:root {
    --title-size: clamp(1.5rem, 5vw, 4rem);
    --h2-size: clamp(1.25rem, 3.5vw, 2.5rem);
    --body-size: clamp(0.75rem, 1.5vw, 1.125rem);
    --slide-padding: clamp(1rem, 4vw, 4rem);
    --content-gap: clamp(0.5rem, 2vw, 2rem);
}

img, .image-container { max-width: 100%; max-height: min(50vh, 400px); object-fit: contain; }
.card, .container { max-width: min(90vw, 1000px); max-height: min(80vh, 700px); }
.grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(min(100%, 220px), 1fr)); gap: clamp(0.5rem, 1.5vw, 1rem); }

@media (max-height: 700px) { :root { --slide-padding: clamp(0.75rem, 3vw, 2rem); --title-size: clamp(1.25rem, 4.5vw, 2.5rem); } }
@media (max-height: 600px) { :root { --title-size: clamp(1.1rem, 4vw, 2rem); --body-size: clamp(0.7rem, 1.2vw, 0.95rem); } .nav-dots, .keyboard-hint { display: none; } }
@media (max-height: 500px) { :root { --title-size: clamp(1rem, 3.5vw, 1.5rem); --slide-padding: clamp(0.4rem, 2vw, 1rem); } }
@media (max-width: 600px) { .grid { grid-template-columns: 1fr; } }
@media (prefers-reduced-motion: reduce) { *, *::before, *::after { animation-duration: 0.01ms !important; transition-duration: 0.2s !important; } }

⚠️ Never negate CSS functions directly. Use calc(-1 * clamp(...)) not -clamp(...).

Phase 1: Content Discovery (New Presentations)

Ask the user (via normal conversation — ask all at once):

1. Purpose: pitch deck / teaching / conference talk / internal presentation?
2. Length: ~5–10 / 10–20 / 20+ slides?
3. Content: ready content, rough notes, or just a topic?
4. Images: any images to include? (provide path like ./assets or ~/Desktop/screenshots)
5. Inline editing: should text be editable in-browser after generation?

If they have content, ask them to share it.

Image Evaluation (if images provided)

1. ls the folder — find all .png/.jpg/.jpeg/.gif/.svg/.webp
2. View each image (Claude is multimodal — use read tool)
3. Mark each as USABLE or NOT USABLE (reason: blurry, irrelevant, etc.)
4. Note what each represents (logo, screenshot, chart, etc.) and dominant colors
5. Build slide outline that incorporates usable images as visual anchors
6. Present outline + image assignments to user for confirmation

If user says no images → skip image pipeline, use CSS-generated visuals (gradients, shapes, patterns) throughout.

Phase 2: Style Discovery

Option A — Direct Selection (user knows what they want)

Pick from presets:

Skip to Phase 3.

Option B — Visual Discovery (default for undecided users)

Ask: What feeling should the audience have?

• Impressed/Confident → Bold Signal, Electric Studio, Dark Botanical
• Excited/Energized → Creative Voltage, Neon Cyber, Split Pastel
• Calm/Focused → Notebook Tabs, Paper & Ink, Swiss Modern
• Inspired/Moved → Dark Botanical, Vintage Editorial, Pastel Geometry

Generate 3 mini style preview HTML files in .tmp-slide-previews/ — each is a single title slide showing typography, colors, and animation. Tell the user the file paths and ask which they prefer.

Never use: purple gradients on white, Inter/Roboto/system fonts, generic blue primary colors.

Use instead: distinctive font pairings (Clash Display, Satoshi, Cormorant, DM Sans), cohesive personality-driven palettes, atmospheric backgrounds.

For full preset specs (colors, fonts, signature elements): read references/STYLE_PRESETS.md.

Phase 3: Generate Presentation

Image Processing (skip if no images)

If user provided images, process with Python Pillow before generating HTML:

from PIL import Image, ImageDraw

# Circular crop (for logos)
def crop_circle(input_path, output_path):
    img = Image.open(input_path).convert('RGBA')
    w, h = img.size
    size = min(w, h)
    img = img.crop(((w-size)//2, (h-size)//2, (w+size)//2, (h+size)//2))
    mask = Image.new('L', (size, size), 0)
    ImageDraw.Draw(mask).ellipse([0, 0, size, size], fill=255)
    img.putalpha(mask)
    img.save(output_path, 'PNG')

# Resize large images
def resize_max(input_path, output_path, max_dim=1200):
    img = Image.open(input_path)
    img.thumbnail((max_dim, max_dim), Image.LANCZOS)
    img.save(output_path, quality=85)

Install if needed: pip install Pillow

Use direct file paths in HTML (not base64), e.g. src="assets/logo_round.png". Never overwrite originals — save processed files with _processed suffix.

HTML Architecture

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Presentation Title</title>
    <!-- Fonts via Fontshare or Google Fonts -->
    <style>
        /* CSS Custom Properties (theme) */
        :root { /* colors, typography with clamp(), spacing */ }
        /* Mandatory viewport base CSS (see above) */
        /* Slide-specific styles */
        /* Animations (.reveal → .visible .reveal) */
        /* Responsive breakpoints */
    </style>
</head>
<body>
    <div class="progress-bar"></div>
    <nav class="nav-dots"></nav>

    <section class="slide title-slide">
        <h1 class="reveal">Title</h1>
        <p class="reveal">Subtitle</p>
    </section>
    <!-- More slides... -->

    <script>
        class SlidePresentation {
            // Keyboard nav (arrows, space)
            // Touch/swipe support
            // Mouse wheel navigation
            // Progress bar + nav dots
            // Intersection Observer for .reveal animations
        }
        new SlidePresentation();
    </script>
</body>
</html>

Required JS Features

• Keyboard navigation: arrow keys, space
• Touch/swipe support
• Mouse wheel navigation
• Progress bar + navigation dots
• Intersection Observer → adds .visible class → triggers CSS animations
• Staggered reveal delays (nth-child 0.1s, 0.2s, 0.3s...)

Inline Editing (only if user opted in)

⚠️ Do NOT use CSS ~ sibling selector for hover-based show/hide — pointer-events: none breaks the hover chain. Use JS with delay timeout:

// Hotzone hover with 400ms grace period
const hotzone = document.querySelector('.edit-hotzone');
const editToggle = document.getElementById('editToggle');
let hideTimeout = null;

hotzone.addEventListener('mouseenter', () => { clearTimeout(hideTimeout); editToggle.classList.add('show'); });
hotzone.addEventListener('mouseleave', () => { hideTimeout = setTimeout(() => { if (!editor.isActive) editToggle.classList.remove('show'); }, 400); });
editToggle.addEventListener('mouseenter', () => clearTimeout(hideTimeout));
editToggle.addEventListener('mouseleave', () => { hideTimeout = setTimeout(() => { if (!editor.isActive) editToggle.classList.remove('show'); }, 400); });

// Keyboard: E key (skip when editing)
document.addEventListener('keydown', (e) => {
    if ((e.key === 'e' || e.key === 'E') && !e.target.getAttribute('contenteditable')) editor.toggleEditMode();
});

Features: contenteditable text, auto-save to localStorage, export/save file.

Code Quality

• Comment every major CSS and JS section
• Semantic HTML: <section>, <nav>, <main>
• ARIA labels on interactive elements
• @media (prefers-reduced-motion: reduce) support

Phase 4: PPT Conversion

from pptx import Presentation
import os

def extract_pptx(file_path, output_dir):
    prs = Presentation(file_path)
    assets_dir = os.path.join(output_dir, 'assets')
    os.makedirs(assets_dir, exist_ok=True)
    slides_data = []

    for i, slide in enumerate(prs.slides):
        data = {'number': i+1, 'title': '', 'content': [], 'images': [], 'notes': ''}
        for shape in slide.shapes:
            if shape.has_text_frame:
                if shape == slide.shapes.title:
                    data['title'] = shape.text
                else:
                    data['content'].append({'type': 'text', 'content': shape.text})
            if shape.shape_type == 13:  # Picture
                img = shape.image
                name = f"slide{i+1}_img{len(data['images'])+1}.{img.ext}"
                path = os.path.join(assets_dir, name)
                with open(path, 'wb') as f: f.write(img.blob)
                data['images'].append({'path': f"assets/{name}"})
        if slide.has_notes_slide:
            data['notes'] = slide.notes_slide.notes_text_frame.text
        slides_data.append(data)
    return slides_data

Install: pip install python-pptx

After extraction: show user the slide structure for confirmation, then proceed to Phase 2 (style), then Phase 3 (generate).

Phase 5: Delivery

1. Delete .tmp-slide-previews/ if created
2. Tell user the output file path
3. Provide navigation instructions:
   • Arrow keys / Space to navigate
   • Scroll or swipe
   • Click nav dots to jump
4. Mention how to customize: :root CSS variables for colors/fonts
5. If inline editing enabled: hover top-left corner or press E
6. Ask if any adjustments needed

Style → Feeling Quick Reference