Lumen inspects DOM elements, explains errors, generates code fixes, and runs them directly on your page — without ever leaving Chrome DevTools.
Lumen is built for developers who want AI help without switching context.
AI generates a fix → click Run → it executes directly on the inspected page and returns the result. No copy-paste ever again.
Select any element in DevTools. Lumen gets its HTML, computed styles, layout, and event listeners — and answers questions about it.
console.error, uncaught exceptions, promise rejections — captured automatically. Click to get a diagnosis and fix in seconds.
Full multi-turn conversations with memory. Multiple sessions, search, all synced to your account across devices.
Switch between Gemini 2.5 Pro, GPT-4o, and Grok 3 with one click. Use the best model for each task.
API keys never touch your browser. All requests go through our secure proxy. You sign in with email — that's it.
Add Lumen from the Chrome Web Store. It takes 30 seconds and doesn't require any API keys or setup.
Press F12 on any webpage. Find the Lumen tab in the DevTools panel. Sign in or create an account in seconds.
Select an element, attach a recent error, or just ask. Lumen responds with specific, actionable answers — and can run the fix directly.
Start free. Upgrade when you're ready. Cancel anytime.
Secure payments via Stripe. Cancel anytime. Questions? support@uselumen.com
When Lumen returns a JavaScript snippet, a green Run on page button appears. Click it and the code executes in the context of the inspected webpage — not in a sandbox. You see the result immediately in the chat. This is unique to browser DevTools extensions and not possible with regular AI tools.
No. Lumen uses its own API keys on the backend. You just sign in with email and you're ready to go. Free users get 10 requests/day, Pro users get 500.
Chrome and all Chromium-based browsers (Edge, Brave, Arc). Firefox support is on the roadmap.
No. Chat messages are stored locally in your browser. Page context (element HTML, error messages) is sent to the AI model only for the duration of that request and is not logged or stored by us.
Yes, at any time. Go to your account settings and cancel — you'll keep Pro access until the end of the billing period. No questions asked.
Gemini 2.5 Flash (default, fast), Gemini 2.5 Pro (highest quality), GPT-4o, GPT-4o mini, o4-mini, Grok 3, and Grok 3 Fast. You switch models with one click in the composer.
Lumen brings AI where your code actually is.
Get started free