Runs & debugging

The dev loop lives on the canvas. Every run is recorded so you can see exactly what each step received and returned.

Run history

Each run stores a record and a per-step record — input, output, status, timing, and any error. Open a run to walk it step by step and find where it went wrong.

Pinned data & partial runs

Pin sample data onto a node so downstream steps have something to work with, then run from any node — you don't have to fire the whole flow to test the middle. You can also fire a single node in isolation.

Retries & replay

• Per-step retries (max tries, wait-between) on any node.
• Replay a whole run from any point with the pinned or modified data.
• A run-level auto-replay policy re-drives runs that failed on transient errors.