Runs
Inspect every execution — status, per-node logs, retries, and why a run is waiting.
Workflows is in Beta. The UI and API may change. The sidebar entry is currently disabled for most workspaces — your account team enables it for pilot orgs.
Every workflow execution is a run. Open the Runs tab on a workflow to see its history. Click a run to open the detail view with per-node information.
Statuses
| Status | Meaning |
|---|---|
| Queued | Trigger fired, run is waiting to start. |
| Running | A node is currently executing. |
| Waiting | Paused for a Wait node, Approval, or external callback. |
| Succeeded | All reachable paths terminated cleanly. |
| Failed | A node errored and there was no recovery path. |
| Cancelled | An operator cancelled the run. |
Per-node detail
Each node row in the run detail shows its status, duration, input snapshot, and output snapshot. Errors include the cause chain — JSONata cause messages are preserved on Transform failures, for example. Tool calls made by Agent nodes and HTTP request/response bodies are visible inline.
Why a run is Waiting
A run enters Waiting in three situations:
- Wait node — an explicit pause for a duration or a named wakeup event. The scheduled resume time is shown in the run detail.
- Approval node — paused until an authorized user approves or rejects. Open the global Approvals view to find and act on pending requests. See Approvals.
- External callback — a long-running integration action is waiting for an upstream service to call back. The run resumes when the callback arrives.
Retries
Failed runs are not automatically retried in Beta. From the run detail page, an operator can re-run the workflow from a failed node. Prior successful node outputs are preserved in the context.
Logs
Logs are scoped to the run — there is no global log surface in Beta. Each per-node log line includes the node id, which correlates it to the node in the graph view.