Features
Everything between your cursor and an answer.
SnapAI is deliberately small: four things, done precisely, on every page you visit.
(a) — In detail
One-click content capture
SnapAI reads the current tab's readable content — headings, body copy, lists and tables — and hands it to your AI with a single click. Nothing touches the clipboard, and you never leave the page.
Screenshot any area
Drag a crosshair over any region of the screen: a chart, a diff, a dashboard panel. The capture goes straight to the model, so visual context works as well as text.
AI-powered analysis
Summarize, translate, explain code, or pull structured data out of a page. Requests go directly from your browser to Google's Generative Language API with your own key.
Seamless integration
Active on every http and https site with active-tab permissions only, plus clipboard write so you can drop the answer wherever you need it.
(b) — Use cases
Researchers
Summarize long papers and compare sources without tab-hopping.
Developers
Explain a stack trace, an API doc section, or a rendered diff.
Content creators
Rewrite, outline, and repurpose a page you're already reading.
Support teams
Snap a customer's screenshot region and get an answer in seconds.
(c) — Settings & API
Your key, your bill
Paste a Google Generative Language API key into the options page. It's stored in local browser storage and used only for your own requests — we never see it.
Permissions, explained
- storage — save your key and preferences locally
- activeTab — read the page you're on, when you ask
- scripting — run the capture script on that page
- clipboardWrite — copy the answer out