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

# Towbar documentation

> Deploy from Git, run services on your own servers, and operate them with a clear history.

Towbar is an open source deployment platform for Ubuntu servers you control.
Define apps and resources in Git, register your servers in Towbar, and follow
each deployment from its source commit to the running service.

## Start with the overview

The homepage pairs the seven-day **Deployments trend** with app, resource, server, and active-incident counts. Click a count to open its list page. Status chips summarize the inventory, and the incident card changes appearance when attention is needed. The recent deployments table shows each workload, environment, ID, status, and request time.

## Start with a working deployment

<CardGroup cols={2}>
  <Card title="Your first deployment" icon="rocket" href="/docs/getting-started">
    Deploy a Dockerfile app and verify its public route, one step at a time.
  </Card>

  <Card title="Install Towbar" icon="server" href="/docs/self-hosting/installation">
    Run the control plane with Docker Compose and create your team and first
    Admin account.
  </Card>
</CardGroup>

New to the model? Read [Core concepts](/docs/concepts) for the difference between
Repositories, servers, workloads, deployments, and releases.

## Build and deploy

<CardGroup cols={3}>
  <Card title="Apps" icon="boxes" href="/docs/apps">
    Configure Dockerfile builds, health checks, and resource allocations.
  </Card>

  <Card title="Resources" icon="database" href="/docs/resources">
    Run eight managed database engines and services from container images.
  </Card>

  <Card title="Preview environments" icon="git-pull-request" href="/docs/previews">
    Review pull requests at stable URLs with separate secrets.
  </Card>
</CardGroup>

## Filter and share views

The secondary sidebar contains the current page’s filters and sections. On mobile, open **Page menu**. Apps, Resources, Servers, and Repositories have **All** and **Needs attention** views with counts for the full inventory.

* **Apps and Resources:** filter by Repository, server, running state, and health. Resources also supports resource type.
* **Servers:** filter by setup status, latest check health, and Scout Agent status. An absent or unfinished check is shown as unknown.
* **Repositories:** filter by latest sync result and whether automatic deployments are enabled or paused.

Needs attention includes workloads with unprepared servers, unhealthy runtime checks, configuration drift, or a missing/stopped container that should be running. Servers need attention when setup is pending or failed, the latest check failed, or an enabled Scout Agent is offline or in error. Repositories need attention after a failed latest sync. An intentionally stopped workload or a disabled Scout Agent does not by itself need attention.

Search and filter selections are stored in the URL. Opening a copied URL restores the view; browser Back and Forward restore earlier selections. Monitoring links preserve the time range, environment, aggregation, and instance selection, while secret-editor links preserve the selected stage and, where applicable, environment. Secret values are never included in these links.

## Operate with context

| I want to…                    | Read                                                                                 |
| ----------------------------- | ------------------------------------------------------------------------------------ |
| Connect a repository          | [Repositories](/docs/repositories)                                                   |
| Add or prepare a host         | [Servers](/docs/servers)                                                             |
| Set environment values        | [Shared secrets](/docs/secrets)                                                      |
| Expose a service securely     | [Domains and TLS](/docs/domains-tls)                                                 |
| Protect database data         | [Backups](/docs/backups) and [Restores](/docs/restores)                              |
| Understand a failed operation | [Health and capacity](/docs/monitoring) and [Troubleshooting](/docs/troubleshooting) |
| Configure Slack or email      | [Notifications](/docs/integrations/notifications)                                    |
| Look up a manifest field      | [Deployment manifest](/docs/reference/deployment-manifest)                           |

## Monitor and investigate

* [Performance and health](/docs/monitoring): select an entity, inspect measurements, and follow operational history.
* [Scout Alerts](/docs/scout-alerts): configure thresholds and HTTP checks, inspect incidents, and verify notification delivery.
* [Deployment comparisons](/docs/deployment-comparisons): compare resource usage over equal observation windows.
* [Vulnerability scanning](/docs/vulnerability-scanning): review image findings and available package fixes.

## Manage your workspace

* [Team settings](/docs/team-settings): manage members, invitations, roles, and team API keys.
* [Personal settings](/docs/personal-settings): update your account, configure sign-in methods, and manage personal API keys.
* [SSH keys](/docs/ssh-keys): store the keys used for server connections.

## What Towbar expects

Towbar is designed for teams that manage their own Ubuntu hosts and trust the
repositories they deploy. Builds run on those hosts. Host administration, DNS,
backup storage, and repository access remain under your control.

Before exposing an installation publicly, review [installation configuration](/docs/reference/environment-variables)
and the [security model](/docs/self-hosting/security). For implementation details, see the
[architecture](/docs/self-hosting/architecture).