Press a shortcut → select anything on screen → ask → get an answer or an action.
No more screenshot, crop, upload, explain. Point lets you draw over any region of any app (images, video, PDFs, remote desktops, sites that block text selection) and hands exactly that region to the AI as context.
Windows: grab the latest .exe (NSIS) or .msi installer from Releases.
- Snip overlay (desktop): Alt+Shift+S anywhere → screen freezes → drag, ask, answer in place
- Select: region, freehand, point, multi-region, and cross-screen selection
- Understand: OCR, layout and table extraction, context-aware explanations
- Act: explain, summarize, translate, copy as a clean format, create a GitHub issue
- Real-Time Lens: action picker, drag/doodle selection, compare
| Part | Tech |
|---|---|
| Desktop shell | Tauri (Rust) — desktop/ |
| UI | React — frontend/ |
| Backend | Python (FastAPI), Docker — backend/ |
| Browser extension | extension/ |
| CI | GitHub Actions builds signed Windows installers on spatial-v* tags |
- How to use it, and what's built: GUIDE.md
- Command reference: COMMANDS.md
See docs/SPEC.md.
Point has no servers and no telemetry. Your screenshots go only to the AI provider you pick, with your own key. See PRIVACY.md.