ScreenSorts

Product Introduction

1. Definition: ScreenSorts is a native macOS application (Apple Silicon optimized) functioning as an offline-first screenshot organizer powered by local AI processing. It utilizes on-device machine learning (Apple Neural Engine) to index, search, and manage screenshots entirely locally.
2. Core Value Proposition: ScreenSorts eliminates the "screenshot graveyard" problem by providing instant offline semantic search, automatic text extraction (OCR), and privacy-first organization for Mac users. Its core promise is "Find Anything. Forget Nothing" without compromising data security.

Main Features

1. Smart Semantic Search:
   • How it works: Uses local vector embeddings generated on-device to enable natural language queries (e.g., "receipt from coffee shop"). Searches pixel content, extracted text, and context. Results are ranked by relevance match percentage.
   • Technology: On-device AI models (Apple Neural Engine), vector database (SQLite).
2. Instant OCR & Metadata Lens:
   • How it works: Performs on-demand Optical Character Recognition (OCR) locally. Detects and extracts text, URLs, IBANs, error codes, etc., with one click. Extracted text is instantly copyable.
   • Technology: Native macOS vision frameworks, local text recognition models.
3. Local AI Tagging:
   • How it works: On-device vision models automatically analyze screenshots and generate semantic tags (e.g., #code, #github). Creates a searchable index without cloud processing.
   • Technology: Apple Core ML models running locally on the Neural Engine.
4. Space Saver & Duplicate Detector:
   • How it works: Applies intelligent compression (3 levels: High Quality, Balanced, Max Compression) to reduce file size by up to 70%. Visual similarity analysis identifies near-identical screenshots, enabling bulk deletion to reclaim storage.
   • Technology: Image compression algorithms, perceptual hashing/computer vision for duplicate detection.
5. Native macOS Integration:
   • How it works: Deep system integration includes menu bar access, native Quick Look (Spacebar preview), Smart Collections (auto-organized by app/date), and system pasteboard compatibility (⌘V pastes screenshot as a file).
   • Technology: macOS Sandbox, native Cocoa APIs, system extensions.

Problems Solved

1. Pain Point: The "Screenshot Graveyard" – inability to find specific screenshots buried in unorganized folders with generic filenames (e.g., Screenshot 2024-05...png). Traditional methods (Finder, manual tagging) are slow and ineffective for visual content.
2. Target Audience:
   • Privacy-Conscious Professionals: Developers, designers, finance/legal/health workers handling sensitive data (code snippets, contracts, medical info, crypto keys).
   • Power Users & Researchers: Individuals collecting references (tweets, charts, designs, articles) who need efficient retrieval.
   • Technical Support & QA Engineers: Those frequently screenshotting terminal errors or UI bugs.
3. Use Cases:
   • Instantly finding a specific error message screenshot in a terminal window.
   • Locating a buried design mockup or color hex code from weeks ago.
   • Retrieving a receipt, contract, or insurance card screenshot without manual tagging.
   • Quickly accessing a tweeted statistic or GitHub repository link captured in a screenshot.
   • Freeing up gigabytes by removing duplicate screenshots.

Unique Advantages

1. Differentiation: Unlike cloud-based alternatives (e.g., Evernote, Google Keep) or basic photo managers, ScreenSorts offers 100% offline operation, zero data upload, and deep semantic understanding powered by local AI, ensuring unparalleled privacy and speed. Competitors lack its integrated OCR, auto-tagging, and duplicate detection without cloud reliance.
2. Key Innovation: The "Zero-Upload Promise" architecture. All AI processing (semantic search, OCR, tagging, duplicate detection) occurs exclusively on the Apple Neural Engine within the macOS Sandbox. Data is encrypted at rest and physically cannot leave the user's device, making it uniquely secure for sensitive information.

Frequently Asked Questions (FAQ)

1. How does ScreenSorts ensure my screenshot privacy?
   ScreenSorts processes all data exclusively on your Mac's Apple Neural Engine within the macOS Sandbox. No screenshots, text, or metadata ever leave your device – no cloud servers, APIs, or internet connection required. Data is encrypted at rest.
2. Does ScreenSorts work without an internet connection?
   Yes, ScreenSorts is designed as an offline-first application. All core features – semantic search, OCR, tagging, duplicate detection – function 100% offline using local AI processing on your Mac's hardware.
3. What are the system requirements for ScreenSorts?
   ScreenSorts requires macOS 13 (Ventura) or later. While it runs on Intel Macs, Apple Silicon (M1/M2/M3 chips) is strongly recommended for optimal performance of the local Neural Engine AI processing.
4. Can I share screenshots organized in ScreenSorts?
   Yes. Simply select a screenshot and press ⌘V to paste it directly into any application (Slack, Figma, Mail) as a standard image file, maintaining your workflow efficiency.
5. How does ScreenSorts' lifetime license work?
   Pay a single fee of $19 USD for a lifetime license, granting use on up to 3 Macs. This includes lifetime free updates to all future versions, with no subscriptions or recurring fees.