Lens

Overview

A single-file, browser-based screen AI assistant. Share your screen, paste any AI provider API key, and a floating chat window lets you ask questions about what's on your screen — powered by live screenshots captured every 3 seconds.

Supports Gemini, OpenAI, Anthropic (Claude), and Groq out of the box.

Getting Started

Prerequisites

• A modern browser (Chrome recommended for screen capture support)
• An API key from one of the supported providers:
  • Gemini — key starts with AIza
  • OpenAI — key starts with sk-
  • Anthropic — key starts with sk-ant-
  • Groq — key starts with gsk_

Installation

No installation needed. Just open lens.html in your browser.

Usage

1. Open lens.html in Chrome
2. Paste your API key — the provider is auto-detected
3. Click Share Screen & Start and select the screen/window to share
4. Click Open Chat Window — a floating popup will appear
5. Move the chat window on top of your shared screen
6. Click Scan & Help or type a question — the AI sees your screen automatically

📁 Project Structure

└── 📄 lens.html   # Entire app: UI, screen capture, and AI calls in one file

License

This project is licensed under the MIT License.