n8n-code-python

To write Python in n8n Code nodes for data transformation, use the standard library only and access input data via patterns like _input.all() or _input.first(), then return results as a list of dictionaries with a json key.

No, you cannot use external Python libraries in n8n Code nodes. The environment enforces a strict no-external-libraries policy, meaning you must build your data transformations using only the built-in Python standard library.

The correct return format for Python code in n8n is a list of dictionaries containing a json key, such as [{"json": {...}}]. Enforcing this safe return format ensures n8n properly processes both single and multi-item outputs.

To access webhook payloads in n8n Python Code nodes, use safe dictionary access patterns like _json["body"]. This ensures secure extraction of incoming webhook data for downstream processing within the standard library constraints.

The main limitations of using Python in n8n workflows are the strict no-external-libraries policy and constrained data-access patterns. You must rely entirely on the standard library and handle errors carefully within the Run Once for All Items or per-item modes.

Use Run Once for All Items mode in n8n Python Code when you need to process batch data transformations across the entire input array at once. This mode allows you to apply standard library logic to multiple items simultaneously before returning the formatted output.