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

# Quickstart

> Get Calseta running locally in under 5 minutes

<Info>
  **Calseta is under active development.** APIs and features may change. We welcome feedback and contributions on [GitHub](https://github.com/calseta/calseta).
</Info>

## Run Calseta locally

Calseta runs as a single stack: FastAPI server, MCP server, and PostgreSQL. Start everything with one command:

```bash theme={null}
docker compose up
```

That's it. The platform starts three services:

| Service        | Port | Description                                          |
| -------------- | ---- | ---------------------------------------------------- |
| FastAPI server | 8000 | REST API for alerts, incidents, enrichment, and more |
| MCP server     | 8001 | Model Context Protocol server for AI agents          |
| PostgreSQL     | 5432 | Primary store and task queue                         |

## Next steps

<CardGroup cols={2}>
  <Card title="How It Works" icon="diagram-project" href="/guides/how-it-works">
    Understand the five-step pipeline: ingest, normalize, enrich, contextualize, and dispatch.
  </Card>

  <Card title="Authentication" icon="key" href="/guides/authentication">
    Create API keys and authenticate your requests.
  </Card>

  <Card title="Alert Sources" icon="bell" href="/guides/alert-sources/overview">
    Connect Microsoft Sentinel, Elastic, Splunk, or a generic webhook.
  </Card>

  <Card title="MCP Setup" icon="plug" href="/mcp-reference/setup">
    Connect your AI agent to Calseta via the MCP server.
  </Card>
</CardGroup>