> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/rowboatlabs/rowboat/llms.txt
> Use this file to discover all available pages before exploring further.

# Desktop App Overview

> Introduction to Rowboat's local-first AI desktop application

Rowboat Desktop is a local-first AI coworker that builds a knowledge graph from your work and helps you act on it. Everything runs privately on your machine - your emails, meetings, and notes stay under your control.

## What is Rowboat Desktop?

Rowboat Desktop connects to your email and meeting notes, builds a long-lived knowledge graph, and uses that context to help you:

* **Draft emails** grounded in past conversations and commitments
* **Prep for meetings** with briefs that include prior decisions and open questions
* **Create presentations** as PDF slides using context from your work
* **Track projects** with memory that compounds over time
* **Record voice memos** that automatically update your knowledge graph

<Note>
  Unlike AI tools that reconstruct context on demand, Rowboat maintains long-lived knowledge that accumulates over time. Context isn't retrieved - it's already there.
</Note>

## Key Benefits

<CardGroup cols={2}>
  <Card title="Private by Default" icon="lock">
    All data stored locally as plain Markdown. No cloud storage required.
  </Card>

  <Card title="Memory That Compounds" icon="brain">
    Context accumulates across emails and meetings, not reconstructed each time.
  </Card>

  <Card title="Bring Your Own Model" icon="key">
    Works with OpenAI, Anthropic, Google, Ollama, and OpenAI-compatible APIs.
  </Card>

  <Card title="Obsidian Compatible" icon="file-lines">
    Knowledge graph stored as Markdown with backlinks you can edit anytime.
  </Card>
</CardGroup>

## How It Works

<Steps>
  <Step title="Connect Your Work">
    Link Gmail, Google Calendar, and meeting note services (Granola, Fireflies, Google Meet).
  </Step>

  <Step title="Build the Knowledge Graph">
    Rowboat extracts people, projects, organizations, and topics into Markdown notes with backlinks.
  </Step>

  <Step title="Get Help with Context">
    Ask Rowboat to draft emails, prep meetings, create docs - all grounded in your accumulated memory.
  </Step>

  <Step title="Inspect and Edit">
    Open the knowledge graph in Obsidian or any text editor. All notes are yours to modify.
  </Step>
</Steps>

## Supported Platforms

* **macOS** - Apple Silicon and Intel
* **Windows** - Windows 10 and later
* **Linux** - AppImage and deb packages

<Note>
  Download the latest version from [rowboatlabs.com/downloads](https://www.rowboatlabs.com/downloads)
</Note>

## Integrations

Rowboat Desktop builds memory from:

* **Gmail** - Email threads and conversations
* **Google Calendar** - Meeting schedules and attendees
* **Google Meet** - Meeting transcripts and recordings
* **Granola** - AI-powered meeting notes
* **Fireflies** - Meeting transcription service

## What Makes Rowboat Different

Most AI assistants start cold every session, searching documents and transcripts to reconstruct context.

**Rowboat maintains long-lived knowledge:**

* Context accumulates over time across all interactions
* Relationships between people, projects, and topics are explicit
* Notes are editable by you, not hidden inside a model
* Everything lives on your machine as plain Markdown files

The result is **memory that compounds**, rather than retrieval that starts from scratch.

## Next Steps

<CardGroup cols={2}>
  <Card title="Explore Features" icon="sparkles" href="/desktop/features">
    Learn about AI assistance, knowledge graph, and background agents
  </Card>

  <Card title="Configure Models" icon="sliders" href="/desktop/model-configuration">
    Set up OpenAI, Anthropic, Google, Ollama, or local models
  </Card>

  <Card title="Understand Workspace" icon="folder" href="/desktop/workspace">
    Explore the \~/.rowboat/ directory structure
  </Card>

  <Card title="Install the App" icon="download" href="https://www.rowboatlabs.com/downloads">
    Download Rowboat for Mac, Windows, or Linux
  </Card>
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.