cofin/railway-tools

Railway Deployment (Flow Tools)

<workflow>

🚀 Official Railway Skills (Highly Recommended)

For full project management, deployment automation, and service orchestration, we highly recommend installing the official Railway agent skills:

• use-railway: Master skill for setting up projects, services, and handling deployments.

Installation:

npx skills add railwayapp/railway-skills

Supplemental Patterns

The patterns below provide additional context for Flow-specific multi-service architectures and serverless constraints.

Serverless / App Sleeping

Railway's serverless feature puts services to sleep after 10 minutes of no outbound traffic.

When to disable serverless (sleepApplication: false):

• Background workers (Celery, SAQ, RQ, Sidekiq)
• Queue processors
• Cron services
• Any service without an HTTP endpoint

Multi-Service Architecture (Web + Worker)

For applications with background task processing, use distinct configuration files (e.g., railway.app.json and railway.worker.json) to manage different runtime requirements.

</workflow> <guardrails> ## Guardrails

• Disable Serverless for Workers: Always set sleepApplication: false for background worker services.
• Use Variable References: Prefer ${{Service.VARIABLE}} syntax (e.g., ${{Postgres.DATABASE_URL}}) instead of hardcoding values.
• Dynamic Port Binding: Never hardcode the port. Always reference ${{PORT}} for application port injection. </guardrails>

<validation> ## Validation

• Verify sleepApplication: false for Workers: Ensure any background worker has sleepApplication explicitly set to false.
• Check PORT Dynamic Reference: Confirm that start commands and environment variables reference ${{PORT}}. </validation>

<example> ## Multi-Service Worker Example

Configuration (railway.worker.json):

{
  "service": {
    "name": "worker",
    "sleepApplication": false
  },
  "deploy": {
    "startCommand": "python -m saq my_app.tasks.worker --workers 4"
  }
}

</example>