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

# Introduction

> Run a team of AI agents that work autonomously, so you don't have to.

# Your AI Team, Always On

Dash lets anyone run a team of AI agents that work autonomously — handling tasks, making decisions, and getting things done — so you don't have to.

Whether you're a solo operator or running a team, Dash puts a crew of AI agents to work for you. Define what they do, set them loose, and check back on results.

<CardGroup cols={2}>
  <Card title="Delegate work" icon="hand-wave">
    Put your AI agents to work. They run tasks, make decisions, and keep going — no hand-holding required.
  </Card>

  <Card title="Any AI provider" icon="microchip">
    Connect to Anthropic, OpenAI, Google, or any major LLM provider. Mix and match across your team.
  </Card>

  <Card title="Desktop" icon="plug">
    Manage your agent team from a desktop app. Monitor activity, chat directly, and deploy new teammates in minutes.
  </Card>

  <Card title="Private by default" icon="lock">
    Everything runs on your machine. Your data stays yours — the only outbound traffic is to your AI provider.
  </Card>

  <Card title="Reach your squad anywhere" icon="users">
    Chat with your squad through Telegram, the built-in interface, a web browser, or integrate via WebSocket API.
  </Card>

  <Card title="Safe by default" icon="shield">
    Secrets encrypted at rest. Squad members sandboxed. Access controlled. Safe to experiment from day one.
  </Card>

  <Card title="Projects" icon="kanban">
    Track tasks and longer-term work in a built-in board. Squad members can create and update tasks, and the Inbox surfaces anything waiting on you.
  </Card>

  <Card title="Sub-agents" icon="users-gear">
    Let an agent bring in help on demand — helpers take a piece of the work each, run at the same time, and report back for the agent to pull together into one answer.
  </Card>
</CardGroup>

Give an agent a big job and it can bring in help. It starts **sub-agents** — helpers that each take one piece of the work, with their own tools and their own space to think — and pulls their answers into one. You see each helper as a row in the chat: open it to watch what it is doing, type to it directly, or stop it. Helpers can also be sent off to work in the background while you carry on; when one finishes, your agent picks the thread back up on its own and tells you what came of it.

## Learn How It Works

New to Dash? Start with the overview to understand how all the pieces fit together.

<Card title="How Dash Works" icon="map" href="/how-dash-works">
  A complete walkthrough of agents, Desktop, messaging apps, and more.
</Card>

## Get Started

<Steps>
  <Step title="Clone and build">
    ```bash theme={null}
    git clone <repo-url> Dash
    cd Dash
    npm install && npm run build
    ```
  </Step>

  <Step title="Launch Desktop">
    ```bash theme={null}
    npm run mc:dev
    ```

    The setup wizard walks you through connecting an AI provider and deploying your first agent.
  </Step>

  <Step title="Chat with your agent">
    Open the **Chat** tab, start a conversation, and your agent is ready to work.
  </Step>
</Steps>

See [getting started](/getting-started) for the full setup guide.