FEATURES
Everything you need,
nothing you don't
Built for power users. Every feature is keyboard-accessible, respects your privacy, and gets out of your way.
Screenshot AI that actually sees
Press a hotkey to capture any region of your screen. The AI sees the screenshot — code, errors, designs, charts — and answers in context.
- Single hotkey capture (default Ctrl+Shift+S)
- Full screen, region, or active window
- Works across all displays
- Image attached to conversation context
Talk. Get answers.
Listen to your interviewer through the speaker — voice to text appears in real time with no latency. The AI answers in one sentence the moment you toggle off.
- Real-time transcription, zero latency
- Multi-language support
- Perfect for verbal warm-ups and behavioral rounds
- Combine voice + screenshot for richer context
Conversations that remember
Every chat keeps full context. Ask follow-ups freely — the AI remembers what you talked about three messages ago, three days ago.
- Session-based conversation memory
- Smart context window management
- Cross-device sync (server mode)
- Local-only mode for privacy
A floating window that respects you
Frameless overlay window that stays on top of your work. Adjustable opacity. Move it with hotkeys. Hide it instantly.
- Always-on-top, transparent overlay
- Adjustable opacity (20%–100%)
- Move + resize via keyboard
- Single-keystroke show/hide
And so much more
Streaming responses
Tokens appear as they generate — instant feedback at 60fps.
Latest AI models
GPT-5, GPT-5.4, Claude Sonnet 4.6, Claude Opus 4.7.
Real-time speaker mode
Sub-second live transcription of your interviewer.
Custom hotkeys
Bind any action to any key combo.
Built for Windows
Native installer, hidden from taskbar, stealth overlay.
Privacy-first
No screenshots logged. No audio stored.
Markdown + Mermaid
Code blocks, tables, and rendered architecture diagrams.
Cancel anytime
Mid-stream? Hit escape. Done.
Reset chat
Start fresh in a single keystroke.