Execution history

Browse local flow runs, inspect per-node and per-case outcomes, and clear the log.

3 min read

Bandura records every flow run in a local history. Nothing is sent anywhere. It answers the two questions that come up constantly: did this pass yesterday? and what exactly did it return last time?

Browsing runs

Open the bottom panel (⌘J) and pick the History tab. Each row is one run: status (passed/failed), flow name, when it ran, and how long it took.

Run detail

Click a row to open the full record of that run: per-node outcomes, error messages, and the result snapshot as it was at the time. This makes regressions concrete: open the last green run, open the red one, and diff by eye what the API returned differently.

In the development build, data-driven runs also keep a Dataset cases list in History. It shows each completed row’s number, verdict, duration, and failure message. Filter passed or failed cases, search for an exact row number or failure text, change row order, and jump directly to a row or the first failure. Long lists are paged in groups of 50. The summary shows the pass rate and calls out rows left unfinished when a run stops early. The case list itself stores no input row values.

Copy matching CSV and Copy matching JSON copy every case matching the current filter; Copy all CSV ignores the filter. The CSV includes row numbers, verdicts, failure messages, and duration. Recognizable credential patterns in failure text are redacted before it is saved, but arbitrary values echoed by a workflow may remain. Review copied text before sharing it. Older runs have only the final per-node snapshot, so they cannot reconstruct these case outcomes. This is not yet a per-case trace of every node or a side-by-side evaluation table.

The Result tab at the top of the flow editor always shows the latest run’s summary for the open flow; History is where the older ones live.

Clearing history

Need a clean slate? Clear history at the top of the History tab deletes every recorded run. It asks for confirmation first and can’t be undone, though it only touches your local run log, never your flows or any other data.

Where it’s stored

History is a local database inside your Bandura profile on this machine. It isn’t part of the workspace, so it’s never committed to Git and doesn’t travel with the repo. Your test files stay clean, your run log stays yours. The exact file, and what is redacted before it is written, are listed on the security and data-handling page.

Related: Run a flow and debug failures · Performance numbers · Running flows in CI, where the report replaces this panel

Last updated

Looking for something else? All 39 articles are on one page in the Help Center.