AI 服務暫時不可用，以下為來源正文，待恢復後補全翻譯。
Notifications You must be signed in to change notification settings Fork 0 Star 1 BranchesTags Open more actions menu Latest commit History 8 Commits 8 Commits Folders and files NameName Last commit message Last commit date LICENSE LICENSE README.md README.md Screenshot 2026-08-24 175957.png Screenshot 2026-08-24 175957.png manifest.json manifest.json sidepanel.css sidepanel.css sidepanel.html sidepanel.html sidepanel.js sidepanel.js Repository files navigation PageLens is a Chrome extension that lets you ask an AI questions about the page you're currently looking at, summarize it, explain it simply, or flag things worth noticing (fees, fine print, deadlines, data collection, etc). It is strictly read-only. PageLens only reads the visible text of the current tab and sends it to Claude for analysis. It never clicks, types, fills forms, or submits anything on your behalf, unlike most "AI browser agent" extensions, there is no action/automation layer at all, which keeps the trust and risk profile much simpler. How it works Click the PageLens toolbar icon to open the side panel. Add your own Anthropic API key in settings (stored locally via chrome.storage, sent directly from your browser to Anthropic's API — it never passes through any third-party server). Ask a question, or use a quick action (Summarize / Explain simply / What should I watch for?). PageLens reads the current tab's text and headings, sends it with your question to Claude, and shows the answer in the panel. Install (unpacked, for development) Clone or download this repo. Open chrome://extensions in Chrome. Turn on Developer mode (top right). Click Load unpacked and select this project's folder. Pin the PageLens icon and click it to open the side panel. Add your Anthropic API key in the ⚙ settings panel. Project structure `` manifest.json # MV3 manifest — activeTab, scripting, sidePanel, storage only background.js # Opens the side panel on toolbar click sidepanel.html/css # Side panel UI sidepanel.js # Page reading + Anthropic API calls + chat rendering icons/ # Toolbar icons `` Permissions, and why each is needed Permission Why activeTab Read the current tab only when you actively use the extension scripting Inject a read-only text-extraction function into the page sidePanel Show the chat UI in Chrome's side panel storage Save your API key/model choice locally host_permissions: Allow reading any page you choose to ask about No cookies, no webRequest, no ability to modify page content or navigate — the extension physically cannot take actions on a page, only read it. Ideas for extending this Highlight-to-ask: right-click a text selection and ask about just that. Per-page memory: cache summaries so re-visiting a page is instant. Export chat as notes / send to a notes app. Support Firefox/Edge via the WebExtensions API (mostly a manifest tweak). Streaming responses instead of waiting for the full answer. License MIT — do whatever you like with it. Resources Readme MIT license Activity Stars 1 star Watchers 0 watching Forks 0 forks Report repository