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

# Quickstart

> Get Aurora running locally for testing and evaluation in under 5 minutes.

# Quickstart

This guide will help you get Aurora running locally using prebuilt images from GitHub Container Registry (GHCR). This is the fastest way to evaluate Aurora.

<Note>
  This quickstart uses prebuilt images. For development or to build from source, see the [Installation](/installation) guide.
</Note>

## Prerequisites

Before you begin, ensure you have:

* Docker and Docker Compose installed
* An LLM API key from one of:
  * [OpenRouter](https://openrouter.ai/keys) (recommended - supports multiple models)
  * [OpenAI](https://platform.openai.com/api-keys)
  * [Anthropic](https://console.anthropic.com/)
  * [Google AI Studio](https://aistudio.google.com/)

<Info>
  Aurora works without any cloud provider accounts! The LLM API key is the only external requirement.
</Info>

## Installation Steps

<Steps>
  <Step title="Clone the repository">
    ```bash theme={null}
    git clone https://github.com/arvo-ai/aurora.git
    cd aurora
    ```
  </Step>

  <Step title="Initialize configuration">
    Run the initialization script to generate secure secrets automatically:

    ```bash theme={null}
    make init
    ```

    This command:

    * Copies `.env.example` to `.env`
    * Generates secure random secrets for `POSTGRES_PASSWORD`, `FLASK_SECRET_KEY`, `AUTH_SECRET`, and `SEARXNG_SECRET`
    * Prepares your environment for first launch

    <Info>
      The `make init` command is idempotent - it won't overwrite existing secrets if you run it again.
    </Info>
  </Step>

  <Step title="Add your LLM API key">
    Edit `.env` and add your LLM API key:

    ```bash theme={null}
    nano .env  # or use your preferred editor
    ```

    Add one of these keys:

    ```bash .env theme={null}
    # OpenRouter (recommended - supports multiple models)
    OPENROUTER_API_KEY=sk-or-v1-...

    # Or OpenAI
    OPENAI_API_KEY=sk-...

    # Or Anthropic
    ANTHROPIC_API_KEY=sk-ant-...

    # Or Google AI
    GOOGLE_AI_API_KEY=...
    ```

    <Note>
      If using OpenRouter, set `LLM_PROVIDER_MODE=openrouter`. For OpenAI, use `LLM_PROVIDER_MODE=openai`.
    </Note>
  </Step>

  <Step title="Start Aurora with prebuilt images">
    Pull and start Aurora using prebuilt images from GHCR:

    ```bash theme={null}
    make prod-prebuilt
    ```

    This command:

    * Pulls the latest Aurora images from GitHub Container Registry
    * Tags them for local use
    * Starts all services with docker-compose

    <Info>
      First launch may take 2-3 minutes as Docker pulls images and initializes services.
    </Info>

    To pin a specific version instead of using latest:

    ```bash theme={null}
    make prod-prebuilt VERSION=v1.2.3
    ```

    Available versions are listed at [github.com/orgs/Arvo-AI/packages](https://github.com/orgs/Arvo-AI/packages).
  </Step>

  <Step title="Get the Vault root token">
    After services start, retrieve the Vault root token from the initialization logs:

    ```bash theme={null}
    docker logs vault-init 2>&1 | grep "Root Token:"
    ```

    You'll see output like:

    ```
    ===================================================
    Vault initialization complete!
    Root Token: hvs.xxxxxxxxxxxxxxxxxxxxxxxxxxxx
    IMPORTANT: Set VAULT_TOKEN=hvs.xxxxxxxxxxxxxxxxxxxxxxxxxxxx in your .env file
               to connect Aurora services to Vault.
    ===================================================
    ```

    Copy the root token value and add it to your `.env` file:

    ```bash theme={null}
    nano .env
    ```

    Add:

    ```bash .env theme={null}
    VAULT_TOKEN=hvs.xxxxxxxxxxxxxxxxxxxxxxxxxxxx
    ```

    <Warning>
      The Vault token is required for Aurora to store and retrieve secrets securely. Without it, cloud connector credentials won't be saved.
    </Warning>
  </Step>

  <Step title="Restart Aurora to load the Vault token">
    Stop and restart Aurora to pick up the Vault token:

    ```bash theme={null}
    make down
    make prod-prebuilt
    ```

    <Info>
      This restart is only needed on first setup. Subsequent restarts aren't necessary unless you change environment variables.
    </Info>
  </Step>
</Steps>

## Access Aurora

That's it! Aurora is now running. Access the web interface:

```
http://localhost:3000
```

### Service Endpoints

* **Frontend**: [http://localhost:3000](http://localhost:3000)
* **Backend API**: [http://localhost:5080](http://localhost:5080)
* **Chatbot WebSocket**: ws\://localhost:5006
* **Vault UI**: [http://localhost:8200](http://localhost:8200)
* **SeaweedFS File Browser**: [http://localhost:8888](http://localhost:8888)
* **Memgraph Lab**: [http://localhost:3001](http://localhost:3001)

## Verify Installation

Check that all services are running:

```bash theme={null}
make logs
```

You should see logs from all services without errors. To view logs for a specific service:

```bash theme={null}
make logs frontend
make logs aurora-server
make logs chatbot
```

## Next Steps

<CardGroup cols={2}>
  <Card title="Configuration" icon="gear" href="/configuration">
    Configure LLM providers, cloud connectors, and integrations
  </Card>

  <Card title="Cloud Connectors" icon="cloud" href="/connectors/overview">
    Add AWS, GCP, Azure, or other cloud provider integrations
  </Card>

  <Card title="Integrations" icon="puzzle-piece" href="/integrations/overview">
    Set up Slack, PagerDuty, GitHub, and other third-party services
  </Card>

  <Card title="Architecture" icon="diagram-project" href="/architecture">
    Learn about Aurora's architecture and components
  </Card>
</CardGroup>

## Build from Source (Alternative)

If you prefer to build images locally instead of using prebuilt images:

```bash theme={null}
make prod-local
```

This builds all images from source and starts the services. Useful for:

* Testing feature branches
* Local development
* Custom modifications

See the [Installation](/installation) guide for detailed development setup.

## Stopping Aurora

To stop all services:

```bash theme={null}
make down
```

This stops and removes all containers but preserves your data in Docker volumes.

## Troubleshooting

<AccordionGroup>
  <Accordion title="Services fail to start">
    Check logs for errors:

    ```bash theme={null}
    make logs
    ```

    Common issues:

    * Missing LLM API key in `.env`
    * Port conflicts (3000, 5080, 5432, 6379, 8080 must be available)
    * Insufficient Docker resources (allocate at least 4GB RAM)
  </Accordion>

  <Accordion title="Vault connection errors">
    Ensure you've:

    1. Retrieved the Vault root token from `vault-init` logs
    2. Added `VAULT_TOKEN` to `.env`
    3. Restarted services with `make down && make prod-prebuilt`
  </Accordion>

  <Accordion title="LLM requests fail">
    Verify:

    * Your API key is valid and has credits
    * `LLM_PROVIDER_MODE` matches your provider (openrouter, openai, anthropic, google)
    * Network connectivity to LLM provider
  </Accordion>

  <Accordion title="Database connection errors">
    Wait for PostgreSQL to fully initialize (check logs with `make logs postgres`). If issues persist:

    ```bash theme={null}
    make down
    docker volume rm aurora_postgres-data
    make prod-prebuilt
    ```
  </Accordion>
</AccordionGroup>

<Warning>
  For production deployments, see the [Production Considerations](/deployment/production-considerations) guide. This quickstart is designed for local testing and evaluation only.
</Warning>