Shipped
What we've built so far
Screen Capture Engine
Region and full-screen recording with hotkey trigger. Lossless local capture pipeline with minimal CPU overhead. The foundation everything else is built on.
Click Tracking
Automatic mouse-click annotation. Every click timestamped and drawn as numbered markers on extracted frames.
Local Whisper Transcription
Whisper.cpp integration for on-device audio transcription. Zero network requests — your voice never leaves the machine.
Markdown Export
Structured markdown bundles with embedded images and narration. Drag and drop into Claude, ChatGPT, or Cursor — it has the full picture instantly.
Multi-AI Format Support
Optimized output for Claude, ChatGPT, Gemini, Copilot, Windsurf, and any markdown-capable AI. One recording, every platform.
Smart Frame Selection
Intelligent frame extraction — skip duplicates, detect UI state changes, and pull only the frames that carry meaning.
Privacy-First Architecture
Full security audit completed. Every byte stays local. No telemetry, no cloud sync, no analytics. Verified by design.
Multi-Monitor Capture
Capture across multiple displays. Region selection works seamlessly across monitor boundaries with per-display DPI awareness.
And more to come.
Keep scrolling for the roadmap
Scroll to explore
What's next
On the horizon
Click any node to explore how every feature connects.