About this project

StudyBar is a free, open-source macOS menu-bar study companion designed to keep your entire study life one click away. It integrates assignments, notes, flashcards, focus timers, schedule, and citations into a single native SwiftUI app (macOS 14+). The app is local-first: no account, no paywall, and no cloud required. Data is stored in a single JSON file on your Mac, with optional iCloud Drive sync. Key features include: - **Overview**: AI-ranked next-up tasks, streak tracking, due-today view, natural-language quick-add, and a "Plan my day" feature that proposes study blocks. - **Assignments**: Due countdowns, checklists, recurring tasks, notifications with Complete/Snooze, AI urgency ranking, and AI "break into steps." - **Notes**: A full Markdown editor with slash menu, tables, checklists, native LaTeX, wikilinks and backlinks, templates, and inline AI for summarizing, rewriting, and proofreading. - **Voice Note**: On-device transcription using Apple Speech or Whisper, with crash-safe autosave. - **Flashcards**: FSRS scheduling, cloze, match and test modes, Anki .apkg/CSV import/export, and AI card generation from your own material. - **Reading**: Track books, chapters, and highlights; attach a PDF to search inside it and ask questions answered from the actual pages with citations (on-device; OCR for scanned PDFs). - **Time & Focus**: Pomodoro (course-linked), stopwatch, focus mode (hide apps + ambient noise), and time blocking. - **Schedule**: Weekly timetable with due markers and exam banners; tap any day to plan it; import classes from .ics feeds or pasted text; unified Calendar (EventKit + iCal feeds + assignments). - **Research & Library**: Citations (APA/MLA/Chicago/IEEE/Harvard/Vancouver/BibTeX), reading list, files, and an RSS/Atom news reader. - **Organize**: Kanban board, courses, and grades. AI is woven into the surfaces you already use, not a separate chatbot. It can organize your material, explain it, and work problems step by step. Inline AI (✨) on any text lets you summarize, rewrite, proofread, or continue; you accept or discard changes. You can attach a PDF and get retrieval-augmented Q&A over relevant pages with citations. A command bar (⌘K) proposes cross-note jobs as confirm-cards. AI runs on Apple's on-device model (macOS 26), Ollama (free local models, qwen2.5:7b recommended), or your own Claude/ChatGPT key (stored in Keychain). On-device and Ollama send nothing off your Mac. System integration includes global hotkeys, a floating command palette, a quick-capture panel, a studybar:// URL scheme, Services menu, Shortcuts.app actions, Reminders push, Canvas import via calendar feed (no API token), undo for deletes, Spotlight indexing, Focus/DND automation, daily backups, and optional iCloud Drive sync. Privacy is a priority: local-first by design, no telemetry, no analytics, no account. Network is used only for features you opt into (Canvas sync, Lookup/News feeds, book covers, link titles, and cloud AI if you choose). Architecture is modular: Core/ contains data models, JSON store, engines and services; Shell/ has the menu-bar status item, root split view, command palette, and shared components; Modules/ has one file per feature. Adding a module is as simple as a view plus a line in Core/ModuleRegistry.swift. Installation is via Homebrew (brew install --cask --no-quarantine hqn07/studybar/studybar), a .dmg download from the latest release, or building from source with XcodeGen. The app is not notarized (requires a paid Apple Developer ID), so macOS may show a one-time warning; right-click → Open or xattr -dr com.apple.quarantine resolves it. Feedback and contributions are welcome. Email is the fastest way to reach the developer; there are also GitHub issues and Discussions. The project is MIT licensed, free to use, modify, and share.