Skip to main content
The Cloud Runner is Reqflo’s managed execution environment for running journeys outside the editor. You point it at a journey, it compiles a build artifact, and workers handle execution — tracking progress, managing retries, and producing structured output. It’s designed for reliability at whatever scale you need, from a single CI check to hundreds of parallel data-driven runs.

​
When to use it

  • You’re running journeys in CI/CD and want runs to complete reliably without depending on a developer’s local machine.
  • You’re scheduling automated runs (nightly regression, scheduled smoke tests).
  • You’re doing data-driven runs across a large dataset and want true parallel execution.
  • You want a persistent, auditable record of run history with full request/response evidence.
  • Your journey uses cloud-managed secrets that should never leave the secure execution environment.

​
Key concepts

Build artifacts. Before a cloud run, the journey is compiled into a build artifact that captures the complete execution plan: steps, value references, assertions, and configuration. The Cloud Runner executes the artifact, not the live journey definition. This makes cloud runs repeatable — the same artifact produces the same run regardless of when you execute it. Durable execution. Cloud runs are durable — if a worker is interrupted, the run can resume or be retried rather than simply disappearing. See Durable execution for the mental model. Secret resolution. When a journey uses secret references, the Cloud Runner resolves them from cloud-managed secrets at execution time. Secret values are never passed through the CLI or displayed anywhere — the runner fetches them securely during the run. Structured output. Cloud runs produce structured results — per-step request/response data, assertion outcomes, derived values, errors, and logs — that are stored and queryable. You can retrieve results with --output flags or via reporting integrations. Parallel data runs. The Cloud Runner is the right execution target for data-driven runs with large datasets, since each row runs as an independent worker job.

​
How it works

1

Compile the artifact

When you trigger a cloud run, Reqflo compiles the current journey into a build artifact. The artifact is stored and associated with the run for auditability.
2

Dispatch to workers

The Cloud Runner receives the artifact and dispatches execution to one or more workers. For data-driven runs, each row dispatches as a separate job.
3

Execute steps durably

Workers execute the artifact’s steps in order, resolving secrets, sending requests, evaluating assertions, and collecting derived values. If a worker is interrupted, durable execution ensures the run can recover.
4

Collect and store results

Results — per-step and overall — are collected, stored, and made available via the Reqflo UI, CLI output flags, and reporting integrations.

​
Examples

Basic cloud run 
reqflo run checkout-flow --env staging --case "Happy path" --cloud
Cloud run with explicit values 
reqflo run checkout-flow --env staging --set user_id=usr_123 --set amount=5000 --cloud
Cloud data-driven run 
reqflo run payment-flow --env staging --data payments.csv --cloud
Cloud run with JUnit output for CI 
reqflo run checkout-flow --env staging --case "Happy path" --cloud \
  --reporter junit --output test-results/
Cloud Runner infrastructure is managed per workspace. See Runners in the admin section for configuration.