Getting started
What is StretchAutomate in one sentence? It's the StretchSuite workflow automation app: an event in one app (a trigger) automatically runs a chain of steps in your other apps.
Do I need to know how to code? No. You build workflows from triggers and steps in the UI. Coding is only relevant if you choose the Webhook/HTTP connector to reach external systems.
What are the six tabs? Dashboard (health), Workflows (your automations), Library (connectors), Connections (linked accounts), Validate & Plan (graph checks), and Run ledger (execution history).
Workflows, triggers, and steps
What trigger types are there? Four: Manual, Webhook, Schedule, and App event (a connector event like lead_created).
What step types are there? Four: action (do something in a connector), filter (a conditional gate), transform (reshape data), and delay (wait). Action is the default.
What are the workflow statuses? Draft, Active, Paused, and Archived. A workflow only fires automatically when it's Active.
Can a workflow have more than one trigger? No — one trigger per workflow. Create separate workflows for separate events.
How do steps run in order? Each step's depends_on list defines the graph. StretchAutomate sorts it into stages (a topological order); steps in the same stage are handled together.
What is a step key and why does it matter? A unique identifier for a step within a workflow. You reference it when mapping data ({{ steps.<key>.field }}) and when setting dependencies. Re‑using a key updates that step in place.
Running and monitoring
What's the difference between simulate and live? Simulate (the default) is a full dry run with no real side effects — it resolves and records every step. Live performs the run for real; action steps are queued for dispatch, so live runs often finalize as partial until dispatch completes.
Will a simulation email my customers or change data? No. Simulate performs no real side effects. It's safe to run repeatedly.
What do the run statuses mean? simulated (clean dry run), partial (some steps didn't fully complete — unresolved references, or actions queued in live mode), and failed (the run hit an error).
Where do I see what actually happened? The Run ledger — it records every run with a per‑step trace of statuses, resolved inputs, and outputs. The Dashboard summarizes overall health.
What does the dashboard's failed count include? Both failed and partial runs, so partial runs stay on your radar.
Connectors, connections, and the engine
What's the difference between a connector and a connection? A connector is a capability (an app with triggers and actions). A connection is a linked account — a named credential reference a workflow uses to reach a connector.
Does a connection store my password? No. Connections store a credential reference, never a plaintext secret.
Why can't I run or publish a workflow? Running, publishing, ad‑hoc step evaluation, and retry computation require the StretchAutomate engine to be enabled. Reading validation and plan still works when it's off (you'll see the last snapshot or a note). Ask your administrator to enable the engine.
What does "Validate" actually check? Whether the graph is structurally sound — returning valid, plus errors (must fix) and warnings (advisory). Publishing validates first and refuses to activate an invalid workflow.
AI and organization scoping
Can StretchAutomate build a workflow for me? With AI suggestions enabled, describe the automation in plain language and get a suggested name, trigger, and steps to refine. It only proposes structure — you still validate and simulate before activating.
Why don't I see AI suggestions? The feature is optional and enabled by an administrator. If it's off, you build manually; nothing else is affected.
Why can I only see some workflows? On the workspace surface you're scoped to your own organization — you only see and manage your own automations. The admin surface spans the platform and requires an authorization step before changes save.
Still stuck?
If a run is failing or behaving unexpectedly, see Troubleshooting Common Problems for a step‑by‑step diagnosis flow, and Retry Policies and Error Handling for recovering partial and failed runs.
Was this helpful?
Help us improve this article
Use these controls to share whether this answer solved the issue. Feedback helps prioritize updates to StretchSuite Support.