Data Pipeline
One input. A sequence of tools. A result you can inspect.
Inputs and results stay in this tab. No ads, analytics, automatic saving or server processing. Maximum 20 steps; input 1 MiB; each output 2 MiB; CSV 10,000 data rows.
1. Input
CSV: comma delimiter, unique nonempty headers, equal column counts. Sorting uses text order (ko), not numeric order. JSON/YAML use JavaScript numbers; high-precision decimals are not guaranteed.
2. Steps
3. Result
Loading local tools…
Cancel stops between steps. It cannot immediately interrupt an existing synchronous tool. A failed run exposes only the last successful partial result.