Privacy-first PDF toolkit - Hash-based Single Page Application with AI-powered Smart Features

🚀 Optimized for performance - 91% smaller bundle with code splitting and lazy loading 🤖 AI-powered analysis - Smart features run 100% locally in your browser

Note: This is version 3.0 - a complete rewrite with modern architecture. For v2 legacy code, see archived repository.

Auto-deploy: Pushes to main branch automatically deploy to production via Vercel.

# Install dependencies
npm install

# Start both app and website
npm run dev:all

# Or start separately
npm run dev          # App-SPA on port 3000
npm run dev:web      # Website on port 4321

• Organize Pages 🤖 - /#organize - Drag & drop page reordering with Smart Analysis (blank pages, duplicates, rotation issues)
• Merge PDF 🤖 - /#merge - Combine multiple PDF files with Smart Merge (date extraction, duplicate detection)
• Split PDF - /#split - Split by structure, page ranges, or file size
• Extract Pages - /#extract-pages - Extract specific pages to new PDF
• Delete Pages - /#delete-pages - Remove unwanted pages
• Rotate PDF - /#rotate - Rotate pages by 90/180/270 degrees

• Add Text PDF - /#add-text - Add custom text with fonts and colors
• Edit Text PDF - /#edit-text - Edit existing text in PDF
• Add Form Fields - /#add-form-fields - Add interactive form fields
• Watermark PDF - /#watermark - Add text/image watermarks
• Sign PDF - /#sign - Add digital signatures
• Flatten PDF - /#flatten - Flatten forms and annotations

• Protect PDF - /#protect - Add password protection (40-256 bit)
• Compress PDF - /#compress - Reduce file size intelligently

• Images to PDF 🤖 - /#images-to-pdf - Convert JPG/PNG to PDF with Smart Image Filter
• PDF to Images - /#pdf-to-images - Export pages as JPG/PNG
• PDF to Word - /#pdf-to-word - Convert PDF to DOCX
• Word to PDF - /#word-to-pdf - Convert DOCX to PDF
• OCR PDF - /#ocr - Extract text with multi-language OCR
• Extract Images - /#extract-images - Extract embedded images

🤖 = AI-powered Smart Features (100% local, no external API calls)

# Development
npm run dev          # App-SPA dev server (port 3000)
npm run dev:web      # Website dev server (port 4321)
npm run dev:all      # Both servers simultaneously

# Production
npm run build        # Build app-spa
npm run build:web    # Build website
npm run build:all    # Build both

# Code Quality
npx tsc --noEmit     # Type check
npm run lint         # Lint code

• 100% client-side — files never leave your device, fully private
• AI-powered Smart Features — automatic analysis runs locally in browser
  • Smart Merge: date extraction, duplicate detection, sort suggestions
  • Smart Organize: blank pages, duplicates, chapters, rotation issues
  • Smart Image Filter: low-quality detection, blur/contrast analysis
• Code splitting — tools load on demand (~74 KB initial, -91% from v2)
• Multi-language — EN, RU, DE, FR, ES with full i18n support
• Dark mode — automatic theme switching with Tailwind CSS
• Hash routing — SEO-friendly, works anywhere without server config
• OCR support — multi-language text extraction with Tesseract.js

├── src/                           # React App-SPA (React 19.1.2 + TypeScript)
│   ├── components/
│   │   ├── tools/                 # 20 PDF tool components (lazy loaded)
│   │   ├── smart/                 # AI-powered UI panels
│   │   └── layout/                # Sidebar, TopBar, etc.
│   ├── services/
│   │   ├── pdfService.ts          # Core PDF operations (pdf-lib, pdfjs-dist)
│   │   ├── smartMergeService.ts   # Smart Merge AI logic
│   │   ├── smartOrganizeService.ts # Smart Organize AI logic
│   │   └── smartImageService.ts   # Smart Image Filter AI logic
│   ├── hooks/                     # useHashRouter, useI18n, useSharedFile
│   ├── locales/                   # 5 language JSON files (EN, RU, DE, FR, ES)
│   └── types/                     # TypeScript definitions
├── website/                       # Astro static site generator
│   ├── src/
│   │   ├── pages/                 # 20+ SEO landing pages
│   │   ├── content/blog/          # Blog articles (MDX)
│   │   └── components/            # Astro components
│   └── public/                    # Static assets
├── src-tauri/                     # Tauri desktop app (optional)
├── vite.config.ts                 # Code splitting + chunk optimization
├── tailwind.config.js             # Custom theme (ocean-*, privacy-*)
└── CLAUDE.md                      # AI assistant development guide

This project contains TWO applications:

• Hash routing: /#merge, /#split, /#organize, etc.
• 100% client-side: All PDF processing in browser (pdf-lib, pdfjs-dist, tesseract.js)
• AI-powered: Smart features with local analysis (no external APIs)
• Code splitting: React.lazy() + manual chunks (~74 KB initial load)
• Multi-language: 5 languages with useI18n hook
• State management: React hooks only (no Redux/Zustand)

• 20+ SEO landing pages: /merge-pdf, /split-pdf, /organize-pdf, etc.
• Schema.org markup: Rich snippets for Google search
• Blog system: MDX articles with automatic sitemap
• Performance: Static HTML, optimized for Core Web Vitals
• Links to App-SPA: Each page has "Go to Tool" button → hash route

User searches "merge pdf"
  ↓
Lands on /merge-pdf (SEO landing page)
  ↓
Clicks "Go to Tool" button
  ↓
Navigates to /#merge (React app loads tool)

• Production: https://localpdf.online
• Repository: https://github.com/ulinycoin/clientpdf-pro
• Development Roadmap: ROADMAP.md
• AI Development Guide: CLAUDE.md
• Issues & Feedback: https://github.com/ulinycoin/clientpdf-pro/issues

Frontend:

• React 19.1.2 + TypeScript + Vite 6
• Tailwind CSS 3 (custom theme)
• Hash-based routing (custom useHashRouter)

PDF Processing:

• pdf-lib + @pdf-lib/fontkit - PDF creation/editing
• pdf-lib-plus-encrypt - Password protection
• pdfjs-dist - PDF rendering and text extraction
• tesseract.js - OCR with multi-language support

SEO & Content:

• Astro 5 - Static site generator
• MDX - Blog articles
• Schema.org - Structured data

AI Features:

• 100% client-side analysis (no external APIs)
• SHA-256 hashing for duplicate detection
• Multi-language pattern recognition (EN, DE, FR, ES, RU)
• Image quality analysis (blur, contrast, sharpness)

Tools: 20 PDF tools (3 with AI) | Languages: 5 | SEO Pages: 20+ | Bundle: ~74 KB gzip Last updated: December 2025