Kekasatori turns videos, papers, and articles into study material on your Mac, transcripts, summaries, flashcards, mind maps, and a tutor that actually reads your sources. Local-first and private.

Download for macOS Join the beta → Free · macOS 15.5+ · Apple Silicon & Intel · notarized by Apple

What it does

Import anything: a YouTube link, an arXiv / PubMed / DOI paper, a web article, or a multi-chapter book pulled in and made studyable.
Instant study packs: transcripts, summaries, structured notes, flashcards, key concepts, and mind maps generated from the source.
A grounded AI tutor: chat with a tutor that answers from the actual document via retrieval (RAG). Use on-device Apple Intelligence, or bring your own key: Claude, GPT, Gemini, DeepSeek, and more via OpenRouter.
Think it through: a built-in Excalidraw canvas, a notebook, and animated (Manim) explainers.
Download or stream: keep media offline in a searchable library, or stream it each item carries its study pack.
Local-first & private: your library, notes, and API keys stay on your Mac. No account, no tracking, no servers.

Stack

SwiftUI + MVVM on SwiftData. On-device generation uses Apple’s Foundation Models; cloud models route through an OpenAI-compatible client with keys held only in the Keychain. Retrieval uses on-device embeddings (VecturaKit). yt-dlp and a universal ffmpeg ship as bundled, hardened-runtime-signed binaries. Auto-updates via Sparkle.

SwiftUI views → ViewModels → Services (Import, Transcript, Study, Tutor/RAG, LLM Router)
                                   ↓
              Foundation Models (on-device) · OpenRouter (BYOK) · yt-dlp + ffmpeg

Distribution

Signed with a Developer ID certificate and notarized by Apple, distributed directly as a .dmg (outside the Mac App Store). You are responsible for ensuring your use of any third-party source complies with that service’s terms and applicable copyright law.

See the Privacy Policy and Terms.