|Docs

Stripe Provisioning

Railway is available as a provider in the Stripe marketplace. Developers can provision Railway infrastructure directly from the Stripe CLI or dashboard using the Agentic Provisioning Protocol (APP).

Getting Started

Install the Stripe CLI and the projects plugin:

Browse available Railway services:

Provision a resource:

View credentials:

Available Services

Plans

ServicePricingDescription
freeFreeTrial plan — 500h compute, 512 MB RAM, 1 GB disk
hobby$5/mo + usageFor individual developers
pro$20/seat/mo + usageFor teams and production workloads

Deployable Services

ServiceCategoryDescription
postgresDatabaseManaged PostgreSQL
redisCacheManaged Redis
mongoDatabaseManaged MongoDB
hostingComputeDeploy a GitHub repo or Docker image
bucketStorageS3-compatible object storage (Tigris)

Deployable services are components of a plan. Provision a plan first, then add deployable services.

Connecting to Databases

After provisioning, credentials are available via stripe projects env --reveal. Use the connection_string to connect from outside Railway:

Connection strings use Railway's public TCP proxy, accessible from anywhere. For lower latency in production, deploy your application on Railway alongside the database to use private networking.

Hosting Configuration

Provision a GitHub repository:

Deploy a specific branch:

Deploy a Docker image:

After provisioning, the service receives a public URL like https://my-app-production.up.railway.app.

Management

Credential Rotation

Resource Removal

Dashboard Access

Each provisioned resource has a Railway dashboard link in its access configuration. Visit the dashboard for advanced management: logs, metrics, scaling, environment variables, and more.

Extended API

Railway provides management endpoints beyond the standard provisioning protocol. These are available to AI agents after provisioning.

Available operations include:

  • Environment variables — set, get, delete variables on hosting services
  • Domain management — generate Railway public domains
  • Scaling — set replica count (1–100)
  • Redeploy / Restart — trigger rebuilds or container restarts
  • Logs — fetch recent log output
  • Configuration updates — change deployed image or branch