Browser extension

Renota for Chrome

An AI reading assistant that remembers what you saved.

Select text on any page and Renota explains it, challenges it, translates it, or connects it to what you read before. Save the parts worth keeping and they become context for every later question — with citations back to the source.

Why it stands out

Five actions on a selection

Understand, analyze, challenge, translate, connect — each with a prompt you can edit, plus any custom buttons you add.

A knowledge base, not a chat log

Saved clips and insights are indexed locally with BM25 and CJK tokenization, so retrieval finds them even when your wording differs.

Writes where you are working

Put the answer straight into the editor you selected from, or copy it. No round trip through another window.

What it does

Selection toolbar

A small toolbar appears next to your selection on any site, with the actions you enabled.

Ask your library

Ask a question and Renota retrieves the relevant clips first, answers from them, and numbers its citations so you can open the exact source.

Bring your own model

OpenAI, DeepSeek, OpenRouter, or any OpenAI-compatible base URL. Test the connection from settings before you rely on it.

Writing back

Replace or insert the answer into the field you came from — useful for replies, notes and drafts.

Page context

The text around your selection travels with the request, which is usually the difference between a generic and a useful answer.

Ten interface languages

Pick a language or follow the browser. English and Simplified Chinese are fully translated.

How to use it

  1. Select and pick an action

    Highlight text, then choose understand, analyze, challenge, translate or connect from the toolbar.

  2. Read with citations

    The answer streams into the side panel. Numbered citations point at the clips and book passages it used.

  3. Save or write back

    Keep the answer as an insight, save the selection as a clip, or insert the answer into the page you are writing in.

What leaves your browser

Only what you send to your model provider, plus optional cloud sync if you turn it on.

  • Page content is sent to your provider only when you run an action, never in the background.
  • Your API key is stored in the extension's local storage and sent only to that provider.
  • Your knowledge base stays in IndexedDB unless you sign in for sync.

Get the extension

Install from the Chrome Web Store, or load the unpacked build from the repository.

Build from source

npm run build, then load the dist folder

Get it

Build from source

npm run build, then load the dist folder

View on GitHub

Requirements: Chrome 120 or newer. An API key from any OpenAI-compatible provider.

Chrome extension questions

Which model providers work?

Anything that speaks the OpenAI chat completions API — OpenAI, DeepSeek, OpenRouter, self-hosted gateways — plus any custom base URL.

Does it work offline?

The extension needs a network connection for model calls. Saving, searching and reading your own clips works locally without one.

Can I edit the prompts?

Yes. Each built-in action has an editable prompt, and you can add custom buttons with their own prompt.

Where is my knowledge base stored?

In the browser's IndexedDB, scoped to the extension. Uninstalling the extension removes it unless you enabled cloud sync.

Does it send anything to your servers?

Only anonymous usage counters, and only if you leave the opt-in telemetry switch on. No page content, no prompts, no API keys.