> ## Documentation Index
> Fetch the complete documentation index at: https://docs.openbunny.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> OpenBunny automatically extracts tasks from your messaging channels using an LLM agent.

OpenBunny is a task extraction and management system that listens to your messaging channels — Slack, WhatsApp, iMessage, Telegram, and Gmail — and uses an LLM agent to automatically create and manage structured tasks.

## How it works

1. **Listen** — OpenBunny connects to your messaging channels and ingests conversations in real time
2. **Process** — An LLM agent analyzes conversation context and decides whether to create, update, or ignore tasks
3. **Manage** — Tasks are organized with priorities, labels, requesters, and reminders
4. **View** — Use the local client, cloud dashboard, or mobile app to manage your tasks

## Components

<CardGroup cols={2}>
  <Card title="Service" icon="server" href="/service/overview">
    The core engine. Connects to channels, runs the LLM agent, and exposes a REST API. Built with Node.js and Hono.
  </Card>

  <Card title="Cloud" icon="cloud" href="/cloud/overview">
    Multi-tenant SaaS backend with a web UI. Powered by Next.js and Supabase with team management and push notifications.
  </Card>

  <Card title="Client" icon="desktop" href="/client/overview">
    Lightweight local web UI with a Kanban board. Connects directly to the service API. No authentication required.
  </Card>

  <Card title="Mobile" icon="mobile">
    Expo React Native app with offline-first sync, push notifications, and team support.
  </Card>
</CardGroup>

## Deployment modes

OpenBunny supports three deployment configurations:

| Mode             | Description                                      | Components               |
| ---------------- | ------------------------------------------------ | ------------------------ |
| **Fully local**  | Everything runs on your machine with SQLite      | Service + Client         |
| **Hybrid**       | Service runs locally, data persists in the cloud | Service + Cloud          |
| **Fully remote** | All components deployed on servers               | Service + Cloud + Mobile |

See [Architecture](/architecture) for a detailed breakdown.