Summarize web pages, ask your PDFs, understand videos and translate whole pages, right in your browser's side panel. No copy-pasting, no tab switching
Can’t open the Chrome Web Store? Download the installer · How to install
Free trial included · Sign in with Google or WeChat · Beta
Get an outline of a YouTube or Bilibili video from its captions. Every point has a timestamp: click it to jump there, then ask follow-up questions.
SideDog reads the article or PDF you are on, so you can ask follow-up questions without switching tabs.
Turn long reports, docs and news into key points, tables or a standalone HTML report in one click.
Read any page in your language, side by side with the original or in its place.
Grab the screen, an area or a single element, and ask about charts, tables and images.
Pages are only sent when you ask a question. Your conversation history stays in your browser.
Chrome: open SideDog in the Chrome Web Store and click Add to Chrome. Microsoft Edge: install it from Microsoft Edge Add-ons. Both update automatically.
chrome://extensions (or edge://extensions) and turn on Developer mode.manifest.json).Installed this way, SideDog does not update automatically. When a new version is out, download the installer here again and replace the folder.
chrome://extensions (or edge://extensions) and turn on Developer mode.manifest.json).Every new account starts with a free trial. Paid plans are coming soon.
SideDog’s answers are generated by Agnes AI’s large language models (agnes-2.5-flash and agnes-3.0-flash).
SideDog is an AI assistant that lives in the side panel of Chrome and Microsoft Edge. It reads the page or PDF you are on, so you can ask questions about it, get a summary or a report, translate the whole page, or capture part of the screen and ask about it.
Desktop Google Chrome and Microsoft Edge, version 114 or later. The same extension works in both.
Every new account starts with a free trial with a daily usage limit. Paid plans will come later, and you will see the price before you pay.
Only what a question needs: your question, the images you attach and, unless you remove it, the text of the current page. Nothing is sent while you simply browse. Conversations are stored in your browser, and there are no ads or tracking.
PDFs with selectable text are read directly. For charts, images and scanned pages, use the screenshot buttons above the input box: capture the screen, drag an area or pick a single element.
The interface, answers and page translation are available in English, Simplified Chinese, Traditional Chinese, Japanese, Korean, French, German, Spanish, Portuguese (Brazil) and Russian. SideDog follows your browser language, and you can change it in settings.
Open the PDF in Chrome or Edge, then turn on "Allow access to file URLs" once: open chrome://extensions (or edge://extensions), click Details under SideDog and switch it on. If it is off, SideDog shows an "Open extension settings" button that takes you there. PDFs opened from a website do not need this.
Yes, for videos with captions (including auto-generated ones). Open the video, then click "Video outline" in the side panel. You get an overview and the main points in order, each with a timestamp you can click to jump to that moment. On Bilibili, some captions are only available after you sign in to Bilibili.