generated: '2026-07-19'
method: searched
source: https://help.launchableinc.com/resources/cli-reference/
note: Launchable publishes no separate sandbox environment, test credentials or magic test values.
  What it does publish is a dry-run mode and a connectivity verification command, plus a hosted web
  console. Nothing here is invented; absent facilities are recorded as published=false.
console:
  url: https://app.launchableinc.com/
  signup: https://app.launchableinc.com/signup
  description: Hosted web app where workspaces, API keys, subsets, trends and test reports are
    managed and inspected.
test_mode:
  published: true
  mechanism: dry-run flag
  flag: --dry-run
  description: Runs any CLI command end to end without sending data to the Launchable service,
    printing what would have been transmitted.
  scope: all CLI commands
verification:
  command: launchable verify
  description: Verifies CLI installation, authentication and connectivity to the service before a
    real run.
observability:
  flag: --log-level audit
  description: Prints the HTTP requests the CLI makes, so a run can be inspected without a sandbox
    tenant.
separate_test_environment:
  published: false
test_credentials:
  published: false
magic_test_values:
  published: false
time_simulation:
  published: false
isolation_guidance:
  mechanism: workspace
  description: Separate workspaces isolate data sets; a scratch workspace is the documented way to
    experiment without polluting production models.
  docs: https://help.launchableinc.com/concepts/workspace/