Message Creator

This node creates Messages, which can include text and/or image content. These messages can be assigned specific roles such as User, AI, or Tool. It is possible to select columns from the input table to generate messages with mixed content, including both text and image data. Additionally, the node supports embedding tool calls within the AI messages. The input table is processed row by row, converting each row into a Message with one or multiple content parts.

Message types:

  • User Message: Contains text and/or image content provided by the user and is assigned the User role. These messages are used to represent user input.
  • AI Message: Contains text and/or image content, assigned the AI role, and may include tool calls. Tool calls specify the tool name, ID, and arguments. These messages are used to represent AI-generated responses or actions, including invoking tools for specific tasks.
  • Tool Message: Contains text and/or image content, assigned the Tool role, and includes the tool call ID. These messages represent outputs or responses from tools invoked.

Options

Role input
Choose whether the message role is provided as a value or as a column.
Role
Specify the role of the message, such as User, AI or Tool.
Role column
Select the column from the input table that contains the message roles.
Name
(Optional) Select the input table column containing the name, or leave as None. Currently only used for the Tool role.
  • Purpose: This column provides a name to differentiate between participants of the same role. In conversations where multiple entities share the same role (such as several users or multiple tools), the name property allows you to uniquely identify each participant.
  • Usage: This is especially useful when simulating structured multi-user dialogues or integrating multiple tools, as it lets the model track who said what and maintain context across turns.
  • Example: In a chat where two users are involved, assigning names like user_A and user_B helps the model respond accurately and refer to the right person or tool, even though both technically have the same role.
Tool call ID column
(Optional) Select the column containing tool call IDs. The ID should match that of the tool call generated by an agent.
Message content
Define the content of the message, including text and/or image parts.
  • Type: Select the type of the message content (Text or Image).
  • Input type: Specify how the content is provided: either directly as a value or from a column in the input table.
  • Image column: Select the column from the input table that contains PNG images.
  • Text value: Enter the text content.
  • Text column: Select the column from the input table that contains text values.
Tool calls
Define tool calls associated with the message, including tool names, IDs, and arguments.
  • Tool name column: Select the input table column containing the tool name.
  • Tool call ID column: Select the input table column containing the tool IDs.
  • Arguments column: Select the input table column containing the tool arguments (JSON string).
Message column name
Name of the output column that will contain the created messages.
Remove input columns
If selected, the columns used to create the new message will be removed from the output table.

Input Ports

Icon
The table containing rows to process. Each row will be converted into a Message.

Output Ports

Icon
The resulting table containing created messages.

Popular Predecessors

  • No recommendations found

Popular Successors

  • No recommendations found

Views

This node has no views

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.

Installation

To use this node in KNIME, install the extension KNIME AI Extension from the below update site following our NodePit Product and Node Installation Guide:

v5.5

A zipped version of the software site can be downloaded here.

Plugin provider: KNIME AG, Zurich, Switzerland
Plugin version: 5.5.0.v202506251500
On NodePit since: 2025-07-02
Last update: 2025-07-25
Tags: Streamable
KNIME versions: Since v5.5

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!