01
Gemini Vision OCR
Images sent as base64 to Gemini 3 Flash for accurate, context-aware text extraction.
02
Auto Title & Tags
Every document gets a descriptive title and 3–5 relevant tags generated by AI.
03
AI Summary
2–3 sentence summary generated for every processed document. Visible on the preview tab.
04
Flashcard Generator
5–10 study flashcards covering key concepts, saved to the database per document.
05
Task Extraction
Pull all tasks, action items, to-dos, and deadlines out of any document in one tap.
06
Formula Extraction
Isolate mathematical equations, scientific notation, and formulas from textbooks.
07
PDF Export
Generates a formatted PDF with title, date, category, tags, summary, and body text.
08
Markdown Export
Exports structured .md file with front matter — title, tags, summary, and full text.
09
AI Chat
Streaming conversation about your document — powered by a ToolLoopAgent on Gemini.
10
Full-Text Search
Search across title, clean text, summary, and tags simultaneously.
11
Category Filters
Notes, Books, Receipts, IDs, Assignments, Whiteboards — filter the library by type.
12
Editable Text
Tap Edit on the Text tab to correct any OCR errors. Changes are saved to the database.
13
Image Persistence
All scans copied to documentDirectory on save — no broken images after restart.
14
Scan Modes
Notes, Book, Whiteboard, Receipt, ID Card — each tunes the AI processing prompt.
15
Animated Scanner
Live scan-line animation in the camera overlay. Corner guide markers for alignment.
16
Turso Database
LibSQL + Drizzle ORM. Documents and flashcards persisted across devices via cloud DB.