annabelle - a simple and clean open-source web co-pilot.

This is not yet on the Chrome web store, will update it when available.

how it works

Screen.Recording.3-25-2024.at.2.13.PM.mp4

Side Panel

options page for manually adding Key

Inference APIs

Currently supports all the inference providers that follows the OpenAI API format such as, Together, Anyscale, etc, and also Google's Gemini. Multi-modality is exclusively supported via Gemini Pro Vision model for now.

Getting Started

First, run the development server:

pnpm dev
# or
npm run dev

Open your browser and load the appropriate development build. For example, if you are developing for the Chrome browser, using manifest v3, use: build/chrome-mv3-dev.

Making production build

Run the following:

pnpm build
# or
npm run build

To use it in a Chromium-based browser (Chrome, Edge, Brave, etc.), enable developer mode on the browser, click on load unpacked, select the build/chrome-mv3-dev folder in your cloned repository, refresh the pages, and start using the co-pilot.