Image Lens

Analyze webpage images and generate editable prompts for image generation.

Image Lens helps you inspect images on web pages and turn them into prompts you can edit, copy, and reuse for image generation. What it does: Adds an analysis button on images you choose on web pages. Generates short and full prompt versions. Supports English and Chinese interface language. Lets you view prompts in a structured format. Lets you edit and copy prompts directly. Optionally generates new images from the prompt with your own API key. How it works: You provide your own API key in the extension settings and can choose Gemini or an OpenAI-compatible provider. When you click an image, the extension reads that image and sends the image data to your selected provider for analysis. On sites that block direct image fetching, the extension locally captures the visible tab and crops only the image area you clicked before sending that cropped image to your selected provider. If image generation is enabled, the extension sends the current prompt to your selected provider to generate images. Privacy model: The developer does not provide or host API keys. Settings are stored locally in chrome.storage.local. The extension does not send your API key or browsing data to the developer. Chinese Detailed Description Image Lens 可以帮你分析网页里的图片，并将图片内容整理成可编辑、可复制、可复用的提示词，方便继续用于生图或二次创作。 主要功能： 在网页图片上提供识别入口。 生成精简版和完整版提示词。 支持中英文界面切换。 支持结构化查看提示词内容。 支持直接编辑和复制提示词。 可选开启生图功能，并使用你自己的 API Key 进行生成。 工作方式： 你需要在插件设置页填写自己的 API Key，并可选择 Gemini 或 OpenAI Compatible 服务。 当你点击网页图片时，插件会读取该图片，并将图片数据发送到你当前选择的服务商进行识别分析。 如果某些网站限制直接读取图片，插件会先在本地截取当前可见标签页，再只裁剪出你点击的图片区域后发送到你当前选择的服务商。 如果开启了生图功能，插件会将当前提示词发送到你当前选择的服务商生成图片。 隐私方式： 开发者不会提供或托管 API Key。 插件设置保存在本地 chrome.storage.local。 插件不会把你的 API Key、网页内容或生成结果发送给开发者。 Latest update: Adds optional OpenAI-compatible provider support for both prompt analysis and image generation. You can now configure provider-specific Base URL, API key, and model settings separately for analysis and generation. This update also keeps the existing Gemini workflow, bilingual interface, structured prompt view, and improved fallback handling for websites that block direct image fetching.