Artisan Commands

SKE lets you run artisan commands against any deployed environment. Commands execute inside a fresh container instance with full access to your application’s environment variables, database, and cache.

Running a command

ske command:run "migrate --force" --env production

  Migrating: 2026_09_15_create_orders_table
  Migrated:  2026_09_15_create_orders_table (42ms)

The command runs in a dedicated invocation, isolated from web traffic. On AWS, this creates a separate Lambda invocation. On GCP, it runs as a Cloud Run job.

Tinker

Open an interactive tinker session:

ske tinker --env production

>>> User::count()
=> 1547
>>> Cache::get('stats:daily')
=> ["visits" => 4291, "signups" => 23]

Tinker runs in your deployed environment with the same database, cache, and environment variables.

Timeouts

Commands have a separate timeout from web requests, configurable in ske.yml:

name: my-app
environments:
  production:
    timeout: 30        # Web request timeout (seconds)
    cli_timeout: 120   # CLI command timeout (seconds)
    cli_memory: 1024   # Memory for CLI commands (MB)

Setting	Description	Default
cli_timeout	Maximum execution time for commands	120 seconds
cli_memory	Memory allocation for commands	Same as memory

Long-running commands like imports or data migrations can use higher timeouts. AWS Lambda supports up to 15 minutes; Cloud Run supports up to 60 minutes.

How it works

The SKE adapter handles command execution through the bridge package:

1. The API sends a command payload to the adapter
2. The adapter creates a synthetic POST request to /__ske/cli
3. The bridge package executes the artisan command
4. Output is streamed back through the API to your CLI

Common uses

# Run database migrations
ske command:run "migrate --force" --env production

# Clear caches
ske command:run "cache:clear" --env staging

# Check queue status
ske command:run "queue:monitor" --env production

# Run a custom command
ske command:run "app:import-data --source=s3" --env production