railway-cli

name	railway-cli
description	Manage Railway cloud deployments via the Railway CLI. Use when the user wants to deploy, manage services, set variables, view logs, link projects, add databases, configure domains, manage volumes, or perform any Railway platform operation from the terminal.

Railway CLI

Operate the Railway platform from the command line. Covers the full lifecycle: authentication, project setup, deployment, services, variables, environments, logs, domains, volumes, functions, and local development.

Source of truth: Railway CLI Docs

Handling Missing Installation

Assume railway is installed. If a command fails with "command not found" or similar, install it before retrying:

# macOS (preferred)
brew install railway

# Any OS with Node.js >= 16
npm i -g @railway/cli

# Shell script (macOS / Linux / WSL)
bash <(curl -fsSL cli.new)

After install, verify with railway --version, then resume the original task.

Self-Discovery

The Railway CLI evolves across versions. When you encounter an unfamiliar subcommand or need to check current flags:

railway --help              # Top-level commands
railway <command> --help    # Subcommand flags and usage

Always prefer --help output over memorized flags when composing complex commands. This keeps behavior correct even as the CLI updates.

Authentication

railway login                   # Browser-based login
railway login --browserless     # Token-based login (CI, SSH sessions)
railway logout
railway whoami                  # Verify current identity

CI/CD token auth (no interactive login):

• RAILWAY_TOKEN — project-scoped actions
• RAILWAY_API_TOKEN — account/workspace-scoped actions

RAILWAY_TOKEN=<token> railway up

Project Lifecycle

Create or link

railway init                    # Create new project (interactive)
railway link                    # Link cwd to existing project
railway unlink                  # Unlink cwd

Inspect

railway status                  # Current project/service/environment info
railway list                    # All projects in workspace
railway open                    # Open project in browser

Deployment

railway up                      # Deploy cwd
railway up --detach             # Deploy without tailing logs
railway deploy --template postgres  # Deploy from a template
railway redeploy                # Redeploy latest deployment
railway restart                 # Restart service (no new build)
railway down                    # Remove latest deployment

Services

railway add                     # Add service (interactive)
railway add --database postgres # Add a database (postgres, mysql, redis, mongo)
railway add --repo user/repo    # Add service from GitHub repo
railway service                 # Switch linked service (interactive)
railway scale                   # Scale service replicas
railway delete                  # Delete the project

Variables

railway variable list                   # List all variables
railway variable set KEY=value          # Set a variable
railway variable set K1=v1 K2=v2       # Set multiple at once
railway variable delete KEY             # Remove a variable

Environments

railway environment                     # Switch environment (interactive)
railway environment new <name>          # Create environment
railway environment delete <name>       # Delete environment

Local Development

railway run <cmd>               # Run command with Railway env vars injected
railway shell                   # Open shell with Railway env vars
railway dev                     # Run services locally with Docker

railway run is useful for running migrations, seeds, or scripts that need production/staging credentials without .env files.

Logs & Debugging

railway logs                    # Stream live deployment logs
railway logs --build            # View build logs
railway logs -n 100             # Last N lines
railway ssh                     # SSH into running service container
railway connect                 # Connect to database shell (psql, mysql, redis-cli, mongosh)

Networking

railway domain                  # Generate a Railway subdomain
railway domain example.com      # Attach a custom domain

Volumes

railway volume list
railway volume add
railway volume delete

Functions

railway functions list
railway functions new
railway functions push

Utilities

railway completion bash         # Generate shell completions (bash/zsh/fish)
railway docs                    # Open docs in browser
railway upgrade                 # Upgrade CLI to latest version

Global Flags

These flags work across most commands:

Flag	Description
-s, --service <name-or-id>	Target a specific service
-e, --environment <name-or-id>	Target a specific environment
--json	Output as JSON (useful for scripting)
-y, --yes	Skip confirmation prompts
-h, --help	Show help
-V, --version	Show CLI version

Common Workflows

First-time project setup

railway login
railway init          # or: railway link
railway up
railway domain        # get a public URL
railway logs          # verify deployment

Add a database to an existing project

railway add --database postgres
railway connect       # verify connectivity
railway variable list # check injected DATABASE_URL

Deploy with environment targeting

railway environment new staging
railway up -e staging
railway logs -e staging

Run a one-off command with prod vars

railway run -e production npx prisma migrate deploy

Additional Resources

For the full and most current reference, see reference.md or consult:

• Railway CLI Docs
• railway --help / railway <command> --help
• Railway CLI GitHub