Axon — AI screen assistant, bring your own key

Summarize any page, ask about what's on screen, and autofill forms. Your API key stays in your browser.

Axon adds an AI layer to the page you already have open. Instead of copying text into another site, press a shortcut and ask. Axon reads the current tab — its text, or a screenshot when the answer is visual — and answers in a panel beside it. WHAT IT DOES • Summarize long pages, articles, threads and documentation. • Ask about what's on screen. Axon can transcribe a chart, work through an error trace, or explain a diagram using a vision model. • Explain any selection. Highlight text, click Explain. • Fill forms from details you've saved. Fields are matched by meaning rather than by name, so a field labelled "Mobile number" is filled from a fact you saved as "Phone". It works inside iframes and on forms that React or Vue render after load. • Remember what you tell it. Mention your role or a portfolio link once and the next form already knows. Every stored detail is visible and editable in settings. YOUR KEY STAYS IN YOUR BROWSER Axon has no backend. There is no account, no sign-up, and no Axon server in the request path. You bring an API key from a provider you already trust. Axon keeps it in Chrome's own chrome.storage.sync and sends requests from your browser straight to that provider. Nothing is proxied through us, because there is nothing to proxy through. • Google Gemini — has a generous free tier • Groq • OpenAI • Any OpenAI-compatible endpoint, including Ollama or LM Studio on your own machine, so nothing has to leave your computer at all You can assign a different model to each job: chat, vision, routing and memory. WHAT IT WON'T TOUCH • It never fills passwords, card numbers, CVVs or one-time codes. • It never uses details you saved about other people to fill your own forms. • It asks for no browsing-history permission. Chrome grants it access to a page only for that visit, and takes that access back when you navigate. • No analytics, no telemetry, no third-party scripts. HOW YOU USE IT • Ctrl+Shift+Y — open Axon on the current page • Ctrl+Shift+U — capture the screen and ask about it • Ctrl+Shift+K — open it in Chrome's side panel • Or right-click a selection, an image, or the page The panel can sit in whichever corner suits you, or dock in Chrome's side panel, where it follows the tab you're on and keeps that tab's conversation. Free and open source. Privacy policy: https://uttamgupta2712.github.io/axon/privacy.html