Executions
What it is
An execution is a single run of a process. Every time a process's trigger fires — a webhook arrives, a schedule ticks, an event hits, you click Run in the editor — Hub records the run and its outcome.
Where to find it
There are three ways to reach a run:
- Tenet → Processes → Runs lists every execution across every pipeline, with the workflow, trigger, type, status, and start time. Click a row to open the live view.
- A pipeline's row in the Processes table shows a Runs count; the pipeline editor itself has a Runs panel.
- A Blueprint's edit page has a Runs tab scoped to its own process.
How it works
Live run view. When you click Run in the editor toolbar, Hub takes you straight to a live view of the execution. The page renders the same graph in read-only form and colours each node by its status — pending, running, completed, or failed. The view updates in place until the run finishes.
Inspecting a node. Click any node on the run view to open a right-side inspector showing that step's resolved input, output, scope changes, duration, and any error.
Run context. The toolbar's Run context button opens the inspector seeded with the run itself — the trigger payload that started it, the final scope, and any top-level error — so you can see what fed the run alongside the per-step trace.
Run details. For each run you can see the trigger that fired it, the status (running, waiting, completed, errored), the message history between the agent and any tools it called, and the token usage for the run.