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

# Getting started with HQ

> Sign up, connect a model, and create your first agent — all from your browser in under 2 minutes.

HQ hosted gives you the full platform without managing any infrastructure. You sign up, bring your own model API key, and create agents immediately.

<Info>
  This guide is for the hosted version at [app.yourhq.ai](https://app.yourhq.ai). If you want to self-host HQ on your own machine, see the [self-host quickstart](/getting-started/quickstart) instead.
</Info>

***

## What you need

<CardGroup cols={2}>
  <Card title="An email address" icon="at">
    Used for sign-in (magic link — no password to remember).
  </Card>

  <Card title="A model API key" icon="plug">
    Any provider: OpenAI, Anthropic, or Ollama. You'll connect this during the onboarding wizard.
  </Card>
</CardGroup>

***

## Step 1 — Sign up and complete the wizard

Go to [app.yourhq.ai](https://app.yourhq.ai) and click **Get started**. The onboarding wizard walks you through everything in one flow:

<Steps>
  <Step title="Welcome">
    Enter your **name** (how agents address you) and confirm your **workspace name** (auto-generated from your first name, editable).
  </Step>

  <Step title="Intent">
    Pick what you'll use HQ for: outreach, closing deals, hiring, publishing content, running operations, or just exploring. This seeds your workspace with a matching context preset and recommends your first agent template.
  </Step>

  <Step title="Model provider">
    Choose **Anthropic**, **OpenAI**, or **Ollama** and enter your API key. HQ validates it inline before continuing. Your key is stored securely and never leaves your workspace infrastructure.

    <Tip>
      **Not sure which to start with?** Anthropic Claude Sonnet or OpenAI GPT-4.1 are the most capable general-purpose choices. You can add more providers in Settings → Connections later. [Compare providers →](/guides/connect-model-provider)
    </Tip>
  </Step>

  <Step title="First agent">
    HQ recommends an agent based on your intent (e.g. "Scout" for outreach, "Ghost" for publishing). Customize the name and emoji, then create it.

    Optionally connect a messaging channel:

    * **Telegram** — create a bot at [@BotFather](https://t.me/botfather) and paste the token
    * **Discord** — create a bot in the [Developer Portal](https://discord.com/developers/applications) and paste the token
    * **Slack** — paste App-Level Token and Bot Token from your [Slack app](https://api.slack.com/apps)
    * **Skip** — interact through tasks, routines, and the remote desktop instead
  </Step>
</Steps>

After agent creation, you land on the **Tasks** page with a suggested first task pre-filled based on your intent.

***

## Step 2 — Pair your channel (Telegram / Discord only)

If you connected Telegram or Discord, send a DM to your bot. It replies with a 6-digit pairing code. Paste that code in the agent's detail page → **Pairing Code** → click **Approve**. Your next message triggers the agent.

Slack agents are active immediately — no pairing step needed.

***

## You're set up

Your hosted HQ is running. Here's what to explore next:

<CardGroup cols={2}>
  <Card title="Pick a model per agent" icon="sliders" href="/guides/connect-model-provider#per-agent-model-overrides">
    Assign different models and thinking levels to each agent.
  </Card>

  <Card title="Understand agents" icon="robot" href="/concepts/agents">
    Memory, files, skills, org chart, and how agents actually work.
  </Card>

  <Card title="Set usage budgets" icon="gauge-high" href="/guides/usage-budgets">
    Avoid surprise API bills with per-agent monthly limits.
  </Card>

  <Card title="Manage your account" icon="credit-card" href="/hosted/account-billing">
    Add workspaces, manage billing, or cancel.
  </Card>
</CardGroup>

***

## BYO model — how it works

HQ is model-agnostic with a bring-your-own-key approach. You connect your own API keys and pay the model provider directly — HQ never touches your keys for billing or proxies through our servers.

You can connect multiple providers and assign different models per agent. See [Connect a model provider](/guides/connect-model-provider) for the full provider list and per-agent override options.