Testing & Lifecycle

Because automations take real actions, Helix lets you prove one out safely before it goes live — and gives every automation a clear lifecycle from first draft to running in production.

Test with a dry-run

Before an automation runs for real, do a dry-run. A dry-run walks the whole pipeline — evaluating branches, loops, and transforms exactly as a real run would — but simulates the side-effecting steps instead of performing them. You see the path the run would take and what each action step would do (which tool it would call, with what inputs) without actually sending an email, charging a card, or changing a record.

Use a dry-run to confirm the logic is right and the actions are what you expect, then run it for real with confidence. Dry-runs don’t count against your budget.

Draft → Trial → Active

Every automation moves through a simple lifecycle, and it must be promoted through these stages in order before its stimulus can fire it:

Stage	What it means
Draft	You’re still building. A draft automation won’t start on its own — its stimulus is rejected while it’s in this stage, so run it on demand to test.
Trial	Live but watched — a good stage to run against real triggers while you keep an eye on the results.
Active	Fully in production, running on its stimulus.

Promote an automation forward from its detail page as your confidence grows — under the hood this updates the automation’s status — and pause it back at any time.

Version history

Helix keeps a version history of an automation as you edit it. You can review earlier versions and restore one if a change didn’t work out. New runs use the current version; a run already in flight keeps the version it started on, so editing never disrupts an active run.

Re-run and replay

When a run finishes — or fails — you can re-run it, or replay from a specific step rather than starting over. Replay reuses the earlier steps’ results and picks up from the point you choose, which is the fast way to recover after fixing a step or an input.