---
title: Item List Output Parser
description: Parse a comma-separated list of items from the model output
---

Make an AI node answer with a list of items instead of a paragraph.

## What it does

Item List Output Parser is not a step on its own: it has no output port and never runs by itself. You plug it into the **Output Parser** slot of an AI node, and that node uses it.

The AI node tells the model to answer with comma-separated values, such as `foo, bar, baz`. The parser then splits the reply at the commas and returns a list of text items. The format is fixed: you cannot change the separator.

## When to use it

* Get keywords, tags or topics for a piece of text.
* Ask for a few ideas (titles, subject lines) and handle each one separately.
* Feed a list into [Split Out](/nodes/builtin/datatransformation/splitout/) to get one item per entry.

When you need named fields or numbers, use [Structured Output Parser](/nodes/builtin/ai/aidependencies/outputparser/structuredoutputparser/).

## Inputs and settings

There is nothing to set. Use the **Prompt** of the AI node to say how many items you want and what they should be.

## Outputs

This node has no output port. It plugs into a slot on another node instead of passing items along.

* **Output Parser** slot of [Tools Agent](/nodes/builtin/ai/aiagents/toolsagentnode/), [Q\&A Agent](/nodes/builtin/ai/aiagents/qanode/), [RAG Agent](/nodes/builtin/ai/aiagents/ragnode/), [Basic LLM Chain](/nodes/builtin/ai/aiagents/basicllmchainnode/), [Information Extractor](/nodes/builtin/ai/aiagents/informationextractornode/)
* **Base Parser** slot of [Auto-fixing Output Parser](/nodes/builtin/ai/aidependencies/outputparser/autofixingoutputparser/)

On the node it is plugged into, `response` becomes a list of text, for example `["pricing", "onboarding", "support"]`.

## Dependencies and credentials

None. This node needs no connection and has no slots for other nodes.

## Example workflow

1. [Get All Text](/nodes/extension/getalltext/) reads the article on the current tab, with **Output field name** `text`.
2. [Basic LLM Chain](/nodes/builtin/ai/aiagents/basicllmchainnode/) with **Prompt** `Give 5 keywords for this article: {{ $input.text }}`, and an [OpenAI](/nodes/builtin/ai/aidependencies/llm/openai/) model on its **Model** slot.
3. Item List Output Parser connected to the **Output Parser** slot of Basic LLM Chain.
4. [Split Out](/nodes/builtin/datatransformation/splitout/) with **Field To Split Out** `response` turns the list into one item per keyword.

## Common issues

* **You get one long item instead of several.** The model answered with line breaks or bullets instead of commas. Ask for "a comma-separated list" in the prompt, or use a stronger model.
* **An item contains a comma.** Commas always split items. Use [Structured Output Parser](/nodes/builtin/ai/aidependencies/outputparser/structuredoutputparser/) with a list example such as `{ "items": ["a"] }` instead.

## Related nodes

* [Structured Output Parser](/nodes/builtin/ai/aidependencies/outputparser/structuredoutputparser/): named fields in JSON.
* [Auto-fixing Output Parser](/nodes/builtin/ai/aidependencies/outputparser/autofixingoutputparser/): retry with a model when parsing fails.
* [Split Out](/nodes/builtin/datatransformation/splitout/): one item per list entry.
* [Basic LLM Chain](/nodes/builtin/ai/aiagents/basicllmchainnode/): the simplest node to plug it into.