Output & data

The bottom panel is where a run reports back. It has two tabs with different jobs: Console is the full technical log; Output is your data.

The Console tab#

Every step event gets a row: timestamp, status (running / success / error), the instrument, its duration in milliseconds, and any message. Errors are translated into plain English with hints, and error rows are tinted so they're easy to spot in a long run. Hover any row to copy it; the header buttons copy or clear the whole log.

The Output tab#

The Output tab collects only what Output steps emit — the values you actually care about, without the noise. Values print with their optional label; JSON pretty-prints with syntax highlighting; every value has a copy button. When new output arrives while you're looking at the Console tab, a badge on the Output tab counts what you haven't seen.

Writing files#

Two instruments write to disk:

• Output can export its value as JSON or CSV — give it a filename and pick the format. CSV works when the value is a list of objects (like the result of a list Extract), with the object keys as columns. Export can be gated on a condition, so a scheduled flow only writes when there's something worth writing.
• Screenshot saves PNGs.

By default files land in your documents folder, in a stable location you can find later:

Both instruments also accept a full path if you want the file somewhere specific. In an exported script, relative paths resolve against wherever you run the script instead.