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

# Migrate from Railway

> Move a Railway project, its variables, Postgres, and volumes onto InstaCloud.

Railway and InstaCloud think about deployments the same way, so the concepts map almost one to one. A service becomes a **compute** service, Railway Postgres becomes a **postgres** service, variables become project secrets, a `${{Postgres.DATABASE_URL}}` reference becomes a binding, and a volume becomes `insta compute volume` mounted at `/data`. This guide takes you from a running Railway project to the same app on InstaCloud, with its data. Your coding agent does most of it.

InstaCloud offers:

* **Postgres with real branching.** A branch opens as a copy-on-write clone holding all of the parent's data, at its own connection string, so a feature or an agent task can run migrations nowhere near production.
* **Built for coding agents.** A CLI and an MCP server your agent drives directly, with an approval gate on the actions that matter.
* **Flexible deployments.** A GitHub repo with no Dockerfile needed, your own Dockerfile, or an image from any registry.
* **Managed services alongside.** Postgres, Redis, MySQL, MongoDB, and object storage in one project.
* **Real support.** An active community on [Discord](https://discord.com/invite/MPxwj5xVvW).

## Migration steps

We will migrate [Railway's own Django template](https://github.com/railwayapp-templates/django), a Python service with a Postgres database. Its `railway.json` says how it was built. Abridged here: the real `startCommand` also runs `collectstatic`:

```json theme={null}
{
  "build":  { "builder": "RAILPACK" },
  "deploy": { "startCommand": "python manage.py migrate && gunicorn mysite.wsgi" }
}
```

**One code change first.** InstaCloud hands Postgres to your app as a single `DATABASE_URL`. This template reads the five `PG*` variables Railway injects instead, so point it at the DSN before you start. In Django that is `dj-database-url`, and most other stacks take a DSN directly.

### 1. Point a coding agent at your Railway project

Your agent reads the project straight from Railway's API, so it needs your repo name and a project token, not a checkout. Give it the skill first:

```bash theme={null}
npx -y insta@latest agent setup
```

That installs the `insta` CLI, the `insta` skill, and the MCP server for every coding agent on this machine. The skill is what knows how a Railway project maps onto services here.

### 2. Create a project

```bash theme={null}
insta project create mysite
```

Or click **Create Project** in the [console](https://console.instacloud.com). A new project starts empty.

### 3. Ask your agent to migrate

Create a Railway project token first, under **Project Settings → Tokens**. Railway's CLI picks a project interactively otherwise, and your agent cannot answer a picker.

```text title="Prompt" wrap theme={null}
Migrate my Railway service mysite and its Postgres data to InstaCloud. The repo is railwayapp-templates/django, and my Railway project token is in RAILWAY_TOKEN. Follow the insta skill's migration runbook, and stop before sending traffic to it.
```

It will ask you to approve anything that touches production, and it will hand two jobs back to you, because Railway has no command for either. **Order matters.** If you have a volume, download it *first*, while its service is still running, because Railway refuses to read a volume whose service is stopped. Only then scale every service to zero, workers and cron services included, so nothing is still writing when the database is copied.

Leave the Railway project frozen rather than deleted, so you can go back to it.

## Limitations

A few things are not here yet. [Limitations](/limitations/overview) covers the ones people hit most, and is worth the minute before you commit. Stuck anywhere? Come to [Discord](https://discord.com/invite/MPxwj5xVvW).